mirror of
https://github.com/NVIDIA/Model-Optimizer.git
synced 2026-10-02 03:14:52 +08:00
Commit Graph
Select branches
Hide Pull Requests
1679-add-mbridge-dsitillation-for-puzzletron
Model-Optimizer-Agent
aanoosheh/topk-kd-topp-ghost
add-codex-claude-skill-wrappers
agent/fix-aa-next-openhands-timeout
agent/fix-compressed-kimi-ptq
agent/kv-cache-autoquant-pr
agent/pin-nel-launcher-version
ajrasane/diffusers_cudagraphs
ajrasane/diffusion_autodeploy
ajrasane/dynamo-onnx-export
ajrasane/dynamo_export
ajrasane/efficient_vit
ajrasane/normalize-calibrated-request-contract
ajrasane/nvbug-5612316-fp8-attention-qdq
ajrasane/nvbug-6721556-swin-mixed-dtype-add
ajrasane/nvbug_6034518
ajrasane/ort-patching-capability-split
ajrasane/qwen3-omni-30B
ajrasane/qwen3omni_final
ajrasane/readme_update
ajrasane/reduce-diffusers-test-compute
ajrasane/remove-ensure-default-opset
ajrasane/sparse-quant-trt-example
akuriparambi/fix-aq-score-scaling-20260623
alit/moe_dev
asma/MoE_amax_sync
asma/activation-quant-aware-scale-setting
asma/add_support_for_custom_backend
asma/aq_bug_fix
asma/autoquantize-qkv-self-attn-scoring
asma/beebot-speed-benchmark-20260706
asma/distributed-amax-error-handling
asma/fix-nvbug-6532161-fsdp2-tied-params
asma/fix-nvfp4-multi-gpu-export
asma/fp4-stochastic-rounding
asma/fp8-sweep-offset-range
asma/fsdp2_full_state
asma/gptq-scale-algorithms
asma/laq-algorithm
asma/laq-export-support
asma/laq-int3-int2-scale-learning
asma/layerwise_skip_in_meta
asma/mcore-aq
asma/mcore_mse_fp32
asma/nemotron_aq
asma/quant_backends_initial_support
asma/qwen3.8_fp4
asma/qwen38-day0-local-hessian-recipes
asma/scale_learning
asma/sequential-calibrate
asma/super_nvfp4_recipe
asma/tensorquantizer-random-hadamard
asma/vlm-layerwise-calibration
asma/wmse
auto/bump-uv-lock-main-2026-09-28
autocast_resize_fp16_support
backup/24-sparsity-pre-rebuild
bchislett/develop_stash
bchislett/eagle-optimize-dynamic-seqlen
bchislett/eagle-tensorboard-logs
bchislett/eagle-training-qol-moe-etc
bchislett/gpt-oss-120b-EAGLE3-round5
bchislett/specbench-tweaks
bchislett/specdec_bench_vllm_flags
beebot/nvbug-6311597
beebot/nvbug-6359821
chad/eval-model-card-provenance
chenhany/OMNIML-5091-cell-t0-d7
chenhany/autonomous-runner-carve-out
chenhany/dataset-scripts-and-specdec-readme
chenhany/dflash
chenhany/dflash-deepseek-v4-flash
chenhany/eagle3-qwen06b
chenhany/fix-dflash-yaml
chenhany/fix-nvrx-compat
chenhany/gemma-4-vllm-image-fix
chenhany/mmlu-moe-ep-shard-fix
chenhany/omniml-5352-ptyche-mcp
chenhany/omniml-5438-mcp-mlflow
chenhany/qad-extra-state-fix
chenhany/specdec-bench-phase-1a-provenance-and-s3-upload
chenhany/vlm_sft_dataset
chenhany@nvidia.com/omniml-5492
chenjiel/demo_mxfp4_nvfp4
chenjiel/dsv4_cast
chenjiel/eval-skill-tau3-banking
chenjiel/fastmegatron_fp8
chenjiel/fix_moe_calib
chenjiel/fix_seq
chenjiel/int4
chenjiel/iq-1bit-2bit-formats
chenjiel/iq-export-reuse-cuda-decode
chenjiel/iq-kernel-dedup
chenjiel/moe
chenjiel/nvfp4-activation-calib-study
chenjiel/nvfp4-global-scale-study
chenjiel/qwen3-8b-online-eagle3
chenjiel/recipe-nvfp4-experts-mse-fp8-cast-kv
chenjiel/recipe-nvfp4-experts-mse-fp8-cast-kv-2
chenjiel/refactor_qwen35
chenjiel/trim-hf-ptq-example-tests
cjluo-nv-patch-1
cjluo-nv-patch-2
coderabbitai/chat/ea0c193
coderabbitai/docstrings/93bab27
coderabbitai/utg/93bab27
codex/megatron-embedding-weight-quantization
codex/qwen36-fp8-granularity-study
codex/secure-slurm-env-setup
depth_importance_example
dkorzekwa/anymodel_layer_name_predicates
dkorzekwa/claude_qwen35
dkorzekwa/claude_qwen35_distill
dkorzekwa/claude_qwen35_eval
dkorzekwa/doge_datablend_weights_tuning
dkorzekwa/hf_qad_example_clarity
dkorzekwa/kd-gap-data-blend-tuning
dkorzekwa/puzzletron_claude_skill
dkorzekwa/qad_docs
dla_surgeon_integration
docs/eval-mlflow-verification
docs/ptq-pytorch-base-image
edge_llm_example
eval-skill-multiarch-cu130
export_dedup_bkstp
export_dedup_cjluo
export_dedup_cjluo2
export_dedup_simplify
export_dedup_verify
feat/mbridge-qad-pipeline-fixes
feat/nvfp4-conv3d-export
feat/scoped-calibration-algo-cfg
feature/codex-day0-agent-roles
feature/dasc-state-sparsity
feature/dasc-state-sparsity-complete-layers
feature/dasc-state-sparsity-explicit-sort
feature/dasc-state-sparsity-identity-restore
feature/dasc-state-sparsity-lifecycle-fixes
feature/dasc-state-sparsity-parity-gate
feature/dasc-state-sparsity-review-api
feature/dasc-state-sparsity-review-boundary
feature/dasc-state-sparsity-review-contract
feature/dasc-state-sparsity-review-contract-cleanup
feature/dasc-state-sparsity-review-device-independent
feature/dasc-state-sparsity-review-device-test
feature/dasc-state-sparsity-review-final
feature/dasc-state-sparsity-review-final2
feature/dasc-state-sparsity-review-fixes
feature/dasc-state-sparsity-review-policy-device
feature/dasc-state-sparsity-review-public-api
feature/dasc-state-sparsity-review-real-scalars
feature/dasc-state-sparsity-review-rounding
feature/dasc-state-sparsity-review-rounding-cleanup
feature/dasc-state-sparsity-review-validation
feature/dasc-state-sparsity-review-validation-source
feature/dasc-state-sparsity-storage-contract
feature/dasc-state-sparsity-storage-roundtrip
feature/dasc-state-sparsity-validation-helpers
feature/mcore_mse_mixed_precision
feature/puzzletron_v2
feature/qwen-image-dmd2
feature/qwen-image-dmd2-qat
feature/qwen-image-fastgen-qad
feature/qwen-image-svdquant-nvfp4
feature/runtime_opt_vllm
fix-nemotronh-fused-awq-export
fix-nvfp4-gptq-static-export
fix/dynamo-onnx-export
fix/gemma4-vocab-size-export
fix/hf-ptq-pad-token-overwrite
fix/megatron-bridge-untied-lm-head-quantization
fix/modelopt-mcp-log-fetch
fix/nixl-efa-libfabric
fix/nvfp4-static-global-amax-tp-ep-sync
fix/puzzletron-campaign-reliability
fix/specdec-bench-provenance
fix_auto_cast_io_special_nodes_bug
fridah/4_6_act
fridah/algo-cfg-1-capabilities
fridah/algo-cfg-2-write-mask
fridah/algo-cfg-3-compile
fridah/algo-cfg-4-execute
fridah/blog-single-gpu-ptq
fridah/export-module-split
fridah/fidelity-check
fridah/fix-fp8-sweep
fridah/four-over-six-algorithm
fridah/general-fp8-sweep
fridah/glm-tmp
fridah/glm5.1-tmp
fridah/global-hessian
fridah/global-hessian-cw
fridah/improve-local-hessian
fridah/in_exclude_mod
fridah/k3-layerwise-fused-export
fridah/k3-layerwise-fused-export-prerebase
fridah/layerwise-kv-cache-replay-fix
fridah/layerwise-mse
fridah/layerwise-mse-fix
fridah/layerwise-mtp-support
fridah/local-hessian-export
fridah/moe-gate-up-sync-cleanup
fridah/mse-dss
fridah/quantile-calibrator
fridah/super-awq
gh-pages
gkarch/add_attn_pruning
gkarch/add_lmms_eval
gkarch/add_mamba_support_to_runtime_stats
gkarch/add_moe_to_runtime_stats
gkarch/multi-gpu-vllm-benchmarking
gkarch/nemotron-nano-v3
gkarch/puzzletron_v2-nano-dockerfile
gkarch/puzzletron_v2_llama
gkarch/puzzletron_v2_sync_20260730
gkarch/support_nemotronh
gkarch/sync-main-449a3992
gkarch/test-image-build
haoguo/dflash-cosmos3-launcher-example
haoguo/dflash-finetune-example
haoguo/dflash-fp32-master-weights
haoguo/dflash-rope-fp32
haoguo/dflash-training-perf
haoguo/dsv3-eagle
haoguo/eagle-doc
haoguo/eagle-newtrainer
haoguo/fix-0208
haoguo/fix-dflash-rope-theta-rope-parameters
haoguo/fix-launcher-global-vars
haoguo/fix-restore
haoguo/fixci-0611
haoguo/gemma4-e4b-support
haoguo/hardware_autoquant
haoguo/k25dflash
haoguo/launcher-claudemd-scope
haoguo/m3-dspark
haoguo/mla-eagle-K2
haoguo/model_patches
haoguo/nvbug6304444
haoguo/per-model-recipes
haoguo/qwen3-8b-dspark-online-example
haoguo/qwen3-8b-dspark-streaming-example
haoguo/qwen3vl
haoguo/qwen3vl-mrope
haoguo/sgl-backend
haoguo/skierat-cr2-guide
haoguo/spec-doc
haoguo/spec-mixin
haoguo/spec-mixin-new
haoguo/spec-recipe-import
haoguo/streaming-vllm029-fixes
haoguo/temp-offline-eagle-ci
haoguo/test-offline-ci
haoguo/udpate-dflash-recipe
haoguo/vllm-dumper
haoxiz/autocast-mismatch
haoxiz/env-fix
haoxiz/kv_cache
haoxiz/onnx-ptq-trt-abi
haoxiz/onnx-validation
haoxiz/trt-fix
hthadicherla/add-ort-preload-dll-cudnn
hthadicherla/torch-ptq-onnx
hungyuehc/omniml-4944
hungyuehc/omniml-5051
hungyuehc/omniml-5395
hungyuehc/omniml-5415
hungyuehc/omniml-5823-iq2xs-nemotron-recipe
hungyuehc/omniml-5899-cuda
hungyuehc/omniml-5899-export
hungyuehc/omniml-5899-ggml
hungyuehc/omniml-5899-tests-v2
hungyuehc/q8-0-export-recipes
hungyuehc/q8-0-quantization
iputterman/specdec-04-15
iq2xs-unified-export
jennifchen/config_compat
jennifchen/fix_kv_resume
jennifchen/mcore_autoquant
jennifchen/moe_lora
jennifchen/moe_lora_with_sequentialmlp
jennifchen/super_nvfp4_recipe
jennifchen/te_qad_debug
jingyux/24-sparsity
jingyux/24-sparsity-backup-20260708
jingyux/24-sparsity-test
jingyux/3-4-diffusion
jingyux/block-quant-onnx
jingyux/diffusion-autoquant
jingyux/diffusion-skip-softmax
jingyux/diffusion.export
jingyux/fastvideo-0
jingyux/fixed-ltx2-ootm
jingyux/headsparsity
jingyux/kimi-k2
jingyux/kimi-k2-nvfp4
jingyux/lobcq
jingyux/modelopt-peft-te
jingyux/qwen-image-dmd2-svdquant-peft
jingyux/qwen-image-fastgen-qad
jingyux/skipsoftmax-diffusion
jingyux/svd-export
jingyux/svd-mcore
jingyux/triton_fa_skip_softmax_diffusion
jingyux/triton_fa_skip_softmax_diffusion_exp
jingyux/vllm-calib
jingyux/vsa-diffusion
jingyux/vsa-training
jingyux/wan22-svdquant-export
jrausch/anymodel_lmeval
jrausch/pr1769-review
jrausch/puzzletron-gpu-ci-image-workflow
jrausch/puzzletron-gpu-quality-baseline-v4
jrausch/puzzletron-v2-ci-quality-closeout
jrausch/puzzletron-v2-config-simplification
jrausch/puzzletron-v2-qwen-runtime-equivalence
jrausch/puzzletron-v2-structured-results-reporting
jrausch/qwen35-4b-vlm-competitive-campaign-followup-pr2361
juhi/nemotron_35
juhim/autoquant-recipe
juhim/haq_poc
kaix/2_4_sa
kaix/agent-marketplace
kaix/dasc-state-sparsity
kaix/kaix/sparse_attention_calibration_decode
kaix/linear-attention-ci-cleanup
kaix/linear-attention-decode-first
kaix/linear-attention-decode-triton
kaix/linear-attention-qat-m1
kaix/linear-attention-qat-m2
kaix/linear-attention-qat-m3
kaix/linear-attention-qat-m4
kaix/linear-attention-qat-m5
kaix/linear-attention-vllm
kaix/mask-reuse-01-offline-calibration
kaix/mask-reuse-02-vllm-capture
kaix/mask-reuse-03-topology-selection
kaix/mask-reuse-calib
kaix/mla_opt
kaix/modelopt_agent
kaix/skip_softmax_calib
kaix/skip_softmax_calib_formula
kaix/skip_softmax_triton
kaix/sparse_attention_export
kaix/sparse_attention_llm_eval
kaix/sparse_attention_with_triton_backend
kaix/sparse_attn_integration
kaix/sparse_attn_quant
kaix/sparse_attn_quant_mixed_fp16
kaix/sparse_attn_quant_v2
kaix/splitk_skip_softmax
kaix/triattn
kaix/triton-skip-softmax-calib
kaix/triton_decode_skip_softmax
kaix/vllm_decode_skip_softmax
kaix/vllm_skip_calib
kaix/watersic-kv-cache
kinjal/automodel
kinjal/fix_megatron_vllm_fq_export
kinjal/fix_vllm_env_var
kinjal/fix_vllm_fq
kinjal/grouped_tp
kinjal/latest_vllm
kinjal/skip_kv_tokens
kinjal/sungsooh_parallel-fakequant-export-pr
kinjal/vllm_app
kinjal/vllm_kvquant_n
kinjal/vllm_wrapper
kmorabia/mbridge-glm53-flash
kmorabia/prune-gemma4-e4b
kmorabia/scicode-verbosity-mitigation
lmikaelyan/compress-vlms
lmikaelyan/fix-generation-params-pruning
main
mchochowski/puzzle_v2_qwen3.5_0p8
mchochowski/puzzletron_distillation
mchochowski/test-image-build
mxin/fix-export-megatron
mxin/fp4-new-kernel
mxin/general-real-quant-refit
mxin/kv-boundary-skip
mxin/ltx-distill-v2
mxin/modelopt-real-quant-canonical
mxin/modelopt-real-quant-refactor
mxin/nvfp4-kv-scale1-minimal
mxin/pdd-fastgen-refactor
mxin/recipe-defaults
mxin/rl-nano3
mxin/sglang
mxin/test-pr
mxin/vllm_fix
mxin/yaml-include
northbloom
oncallbot/fix-795
pensieve-intern/OMNIML-4634/hidden-state-dump-support
pensieve-intern/OMNIML-4646/hidden-state-dump-support
pensieve-intern/OMNIML-4646/training-support
pensieve-intern/OMNIML-4681/hidden-state-dump-support
pensieve-intern/OMNIML-4681/synth-support
pensieve-intern/OMNIML-4681/training-support
pensieve-intern/OMNIML-4691/hidden-state-dump-support
pensieve-intern/OMNIML-4691/synth-support
pensieve-intern/OMNIML-4691/training-support
pensieve-intern/OMNIML-4701/synth-support
pensieve-intern/OMNIML-4735/deployment-support
pensieve-intern/OMNIML-4745/synth-support
pensieve-intern/OMNIML-4755/synth-support
pensieve-intern/OMNIML-4872/author-yaml
pensieve-intern/OMNIML-4879/cell-t0-d3
pensieve-intern/OMNIML-4879/cell-t0-d7
pensieve-intern/OMNIML-4879/cell-t1-d3
pensieve-intern/OMNIML-4879/cell-t1-d7
pensieve-intern/OMNIML-4885/cell-t0-d3
pensieve-intern/OMNIML-4885/cell-t0-d7
pensieve-intern/OMNIML-4885/cell-t1-d3
pensieve-intern/OMNIML-4885/cell-t1-d7
pensieve-intern/OMNIML-4885/t0_d3
pensieve-intern/OMNIML-4886/t0_d3
pensieve-intern/OMNIML-4926/cell-t0-d3
pensieve-intern/OMNIML-4926/cell-t0-d7
pensieve-intern/OMNIML-4926/cell-t1-d3
pensieve-intern/OMNIML-4927/t0_d3
pensieve-intern/OMNIML-4928/t0_d7
pensieve-intern/OMNIML-4929/t1_d3
pensieve-intern/OMNIML-4930/t1_d7
pensieve-intern/OMNIML-4961/cell-t0-d7
pensieve-intern/OMNIML-4961/cell-t1-d3
pensieve-intern/OMNIML-4961/cell-t1-d7
pensieve-intern/OMNIML-4962/t0_d3
pensieve-intern/OMNIML-4967/cell-t0-d3
pensieve-intern/OMNIML-4967/cell-t0-d7
pensieve-intern/OMNIML-4967/cell-t1-d3
pensieve-intern/OMNIML-4967/cell-t1-d7
pensieve-intern/OMNIML-5022/cell-t0-d3
pensieve-intern/OMNIML-5022/cell-t1-d7
pensieve-intern/OMNIML-5022/t0_d7
pensieve-intern/OMNIML-5057/cell-t0-d7
pensieve-intern/OMNIML-5081/cell-t0-d7
pensieve-intern/OMNIML-5088/cell-t0-d7
pensieve-intern/OMNIML-5095/cell-t0-d3
pensieve-intern/OMNIML-5095/cell-t0-d7
pensieve-intern/OMNIML-5095/cell-t1-d3
pensieve-intern/OMNIML-5095/cell-t1-d7
pensieve/fix-issue-1041
pensieve/fix-issue-1066
pensieve/fix-issue-5997673
pensieve/fix-issue-5997787
pensieve/fix-issue-5997832
pinedrift/dflash2-capture-fix
pull-request/1016
pull-request/1017
pull-request/1027
pull-request/1037
pull-request/1043
pull-request/1105
pull-request/1118
pull-request/1152
pull-request/1153
pull-request/1171
pull-request/1190
pull-request/1213
pull-request/1214
pull-request/1241
pull-request/1272
pull-request/1297
pull-request/1311
pull-request/1315
pull-request/1333
pull-request/1344
pull-request/1378
pull-request/1391
pull-request/1398
pull-request/1415
pull-request/1431
pull-request/1499
pull-request/1504
pull-request/1507
pull-request/1523
pull-request/1530
pull-request/1541
pull-request/1649
pull-request/1725
pull-request/1767
pull-request/1769
pull-request/1774
pull-request/1775
pull-request/1784
pull-request/1802
pull-request/1807
pull-request/1809
pull-request/1816
pull-request/1825
pull-request/1886
pull-request/1944
pull-request/1986
pull-request/1999
pull-request/2040
pull-request/2077
pull-request/2078
pull-request/2094
pull-request/2138
pull-request/2166
pull-request/2167
pull-request/2195
pull-request/2198
pull-request/2218
pull-request/2223
pull-request/2240
pull-request/2247
pull-request/2248
pull-request/2251
pull-request/2292
pull-request/2313
pull-request/2315
pull-request/2316
pull-request/2329
pull-request/2360
pull-request/2361
pull-request/2372
pull-request/2415
pull-request/2416
pull-request/2418
pull-request/2419
pull-request/2420
pull-request/2425
pull-request/2430
pull-request/2431
pull-request/2440
pull-request/2441
pull-request/2453
pull-request/2455
pull-request/2459
pull-request/2461
pull-request/2465
pull-request/2467
pull-request/2472
pull-request/2476
pull-request/2496
pull-request/2497
pull-request/2508
pull-request/2516
pull-request/2517
pull-request/2534
pull-request/2552
pull-request/2568
pull-request/2569
pull-request/2576
pull-request/2581
pull-request/2582
pull-request/2587
pull-request/2589
pull-request/2594
pull-request/2596
pull-request/2597
pull-request/2602
pull-request/2604
pull-request/2605
pull-request/2606
pull-request/2615
pull-request/2620
pull-request/324
pull-request/402
pull-request/452
pull-request/564
pull-request/590
pull-request/624
pull-request/649
pull-request/6529944-dynamic-nvfp4-trt-compat
pull-request/653
pull-request/655
pull-request/723
pull-request/743
pull-request/752
pull-request/835
pull-request/840
pull-request/864
pull-request/921
pull-request/930
pull-request/948
pull-request/976
pull-request/979
puzletron_v2_dockerfile
puzzletron/mbridge
puzzletron_v2
puzzletron_v2_orch
qad/te-per-expert-1550
quant_rotate
quant_rotate_models
release/0.27.0
release/0.27.0-windows
release/0.29.0
release/0.31.0
release/0.33.0
release/0.33.0-windows
release/0.35.0
release/0.37.0
release/0.39.0
release/0.40.0
release/0.41.0
release/0.42.0
release/0.43.0
release/0.44.0
release/0.45.0
release/0.46.0
release/0.47.0
rislam/cask-output-quant
rohjoshi/14sparsity-exp
rohjoshi/fast-compressed-tensors-load
rohjoshi/scale-learning-blog
rohjoshi/sparse-kernel
rohjoshi/sparse24-plus-skipsoftmax
shengliangx/config_yaml
shengliangx/deprecating-trtllm-model_type
shengliangx/distributed-export
shengliangx/docs-update
shengliangx/error-old-quant-cfg
shengliangx/hf-ptq-recipe-override
shengliangx/model-download-1-hf-checkpoint-utils
shengliangx/model-download-2-hf-exporters
shengliangx/model-download-3-hf-ptq
shengliangx/model-download-4-megatron
shengliangx/model-download-5-llm-sparsity
shengliangx/move-modeling
shengliangx/move-modeling_1
shengliangx/move-modeling_2
shengliangx/move-modeling_3
shengliangx/move-modeling_4
shengliangx/move-modeling_5
shengliangx/nv-modelopt-cli
shengliangx/prebuild-onnx-ext-off-test-clock
shengliangx/pyrefly-tool
shengliangx/quant_grouped_mm
shengliangx/qwen3vlmoe-export
skierat/dflash-cosmos3-launcher-example
skip-softmax-visualgen-export
ssalian/thinkact
ssameni/puzzletron-bypass
svelury/fp8pb-mxfp8-export
svelury/gptq-data-gettr
svelury/gptq-export
svelury/gptq-lite-helper
svelury/gptq-local-hessian
svelury/gptq-vq-f
svelury/save-last-2-layer-inputs
svelury/seq-calib-save-restore
talora/compute_hidden_states_trtllm_fix
talora/eagle3-experiments
talora/specdec-bench-july-26
temp/day0-harbor-all-fixes-20260918
temp/day0-harbor-five-fixes-20260921
temp/day0-harbor-five-fixes-20260923
template/new-nas-mode
weimingc/autoq-active-moe-nvfp4-mse-exp
weimingc/autoquant-group-scoring
weimingc/autoquant-mtp-recipes
weimingc/autoquant-per-element-group-scoring
weimingc/autoquant-per-element-parent-scoring
weimingc/autoquant-per-element-scoring
weimingc/autoquant_qwen36_lcb_study
weimingc/fix-export-names
weimingc/fix_awq_quant
weimingc/fix_int4_export
weimingc/fix_mcore_nvpf4_export_vllm
weimingc/inkling-attn-support
weimingc/mtp-quantization-export
weimingc/omniml-5271
weimingc/support_mcore_exclude_modules
weimingc/vllm_MoEAWQ
weimingc/vllm_fake_quant_for_fp8_attn_math
yeyu/SWA
yeyu/afterany-pipeline-fix
yeyu/auto_regression_parallel_draft
yeyu/debug
yeyu/debug_paralllel_draft
yeyu/dflash-minimax-m3
yeyu/eagle_embedding_optional
yeyu/eagle_medusa
yeyu/fix_bug_in_loss_mask
yeyu/fix_realquant_bug
yeyu/kv_cache
yeyu/medusa_share_lm_head
yeyu/minimax-m2.7-streaming-dflash
yeyu/mtp-export-rules
yeyu/parallel_eagle
yeyu/pard
yeyu/refactor_multi_step_attn_mask_for_arbitrary_step
yeyu/remove_embedding_from_eagle
yeyu/sage-attention-diffusion
yeyu/set_block
yeyu/skipsoftmax-diffusion
yeyu/specdec-e2e-reference-run
yeyu/specdec-skill-sheets-dflash2-lilicorr
yeyu/tidar_quant
yueshen/export-fused-layernorm-te-spec
yueshen/fp8_qkvo-nvfp4_mlp
yueshen/gemma-auto-quant
yueshen/launcher-megatron-bridge
yueshen/megatron_generate_vlm
yueshen/qad-megatron-bridge
zhiyu/agentic-workflows-schemas
zhiyu/ct-int4-to-gptq-int4
zhiyu/fix-config-export-issue-rope-scaling-type
zhiyu/handle-glm-mtp
zhiyu/patch-dsv3-config
#1
#100
#1000
#1001
#1005
#1006
#1007
#1008
#1009
#1010
#1011
#1012
#1013
#1013
#1014
#1015
#1015
#1016
#1016
#1017
#1017
#1018
#1019
#1020
#1021
#1022
#1023
#1024
#1025
#1026
#1027
#1027
#1028
#1029
#1030
#1031
#1033
#1034
#1035
#1036
#1037
#1037
#1038
#1039
#1040
#1042
#1043
#1043
#1044
#1046
#1047
#1048
#1049
#1050
#1050
#1051
#1052
#1053
#1054
#1054
#1056
#1057
#1058
#1059
#1060
#1061
#1062
#1063
#1065
#1067
#1068
#1069
#1070
#1071
#1072
#1073
#1074
#1075
#1076
#1076
#1077
#1078
#1079
#1080
#1081
#1082
#1082
#1083
#1084
#1085
#1086
#1087
#1089
#1091
#1094
#1096
#1097
#1098
#1099
#1100
#1101
#1102
#1103
#1104
#1105
#1105
#1106
#1107
#1108
#1109
#1110
#1111
#1112
#1113
#1114
#1115
#1116
#1117
#1118
#1118
#1119
#1120
#1121
#1122
#1123
#1124
#1125
#1125
#1126
#1127
#1128
#1129
#1129
#1130
#1131
#1132
#1133
#1134
#1135
#1136
#1137
#1138
#1139
#1140
#1141
#1142
#1143
#1144
#1145
#1146
#1148
#1149
#1150
#1151
#1152
#1152
#1153
#1153
#1154
#1155
#1156
#1157
#1157
#1158
#1159
#1160
#1161
#1162
#1163
#1164
#1165
#1166
#1167
#1168
#1169
#1170
#1171
#1171
#1172
#1174
#1175
#1176
#1177
#1177
#1178
#1179
#1180
#1181
#1182
#1183
#1184
#1185
#1187
#1188
#1189
#1190
#1190
#1191
#1192
#1193
#1194
#1194
#1195
#1196
#1197
#1198
#1199
#1200
#1201
#1202
#1203
#1204
#1205
#1206
#1207
#1208
#1209
#1210
#1211
#1213
#1213
#1214
#1214
#1215
#1216
#1217
#1217
#1218
#1219
#1220
#1221
#1222
#1223
#1224
#1225
#1226
#1227
#1228
#1229
#1230
#1230
#1231
#1232
#1233
#1234
#1235
#1236
#1238
#1239
#1240
#1241
#1241
#1242
#1243
#1244
#1245
#1246
#1247
#1248
#1249
#125
#1250
#1251
#1252
#1253
#1254
#1256
#1257
#1258
#1259
#1260
#1261
#1262
#1263
#1264
#1264
#1265
#1266
#1267
#1268
#1269
#127
#1270
#1271
#1272
#1272
#1273
#1274
#1275
#1276
#1277
#1278
#1279
#128
#1280
#1281
#1284
#1284
#1285
#1286
#1287
#1287
#1288
#1289
#129
#1290
#1291
#1292
#1293
#1295
#1296
#1297
#1297
#1298
#1299
#1300
#1301
#1302
#1303
#1304
#1305
#1306
#1307
#1309
#131
#1310
#1311
#1312
#1313
#1314
#1315
#1315
#1316
#1317
#1318
#1320
#1321
#1322
#1323
#1323
#1324
#1325
#1326
#1327
#1328
#1329
#1330
#1331
#1332
#1333
#1333
#1334
#1335
#1337
#1338
#1338
#1339
#1340
#1341
#1342
#1343
#1344
#1344
#1345
#1347
#1348
#1349
#1350
#1351
#1352
#1353
#1354
#1355
#1356
#1357
#1358
#1359
#1360
#1361
#1362
#1363
#1364
#1364
#1365
#1367
#1367
#1368
#1369
#137
#1370
#1371
#1372
#1373
#1374
#1375
#1376
#1377
#1378
#1378
#1379
#1380
#1381
#1382
#1383
#1384
#1385
#1386
#1387
#1388
#1389
#1390
#1391
#1391
#1392
#1393
#1394
#1395
#1397
#1398
#1398
#1399
#1400
#1401
#1402
#1403
#1404
#1405
#1406
#1406
#1407
#1408
#1409
#1410
#1411
#1412
#1412
#1413
#1413
#1414
#1415
#1415
#1416
#1417
#1418
#1419
#1420
#1421
#1422
#1423
#1425
#1426
#1427
#1428
#1429
#1430
#1431
#1431
#1432
#1433
#1434
#1435
#1436
#1437
#1438
#1439
#1440
#1441
#1442
#1443
#1444
#1445
#1446
#1447
#1448
#1449
#1450
#1451
#1452
#1453
#1454
#1455
#1456
#1457
#1458
#1459
#1460
#1461
#1462
#1463
#1464
#1465
#1465
#1466
#1467
#1467
#1468
#1469
#1470
#1471
#1472
#1473
#1474
#1474
#1475
#1475
#1476
#1477
#1478
#1479
#1480
#1481
#1482
#1483
#1484
#1485
#1486
#1487
#1488
#1489
#1490
#1491
#1491
#1492
#1493
#1494
#1495
#1496
#1497
#1498
#1499
#1499
#1500
#1501
#1502
#1503
#1504
#1504
#1505
#1505
#1506
#1507
#1507
#1508
#1509
#1511
#1512
#1513
#1514
#1515
#1516
#1517
#1518
#1519
#1520
#1520
#1521
#1522
#1523
#1523
#1524
#1525
#1526
#1526
#1527
#1528
#1529
#1530
#1530
#1531
#1532
#1535
#1536
#1537
#1538
#1538
#1540
#1540
#1541
#1541
#1542
#1543
#1544
#1545
#1545
#1546
#1547
#1548
#1549
#1549
#155
#1550
#1551
#1551
#1553
#1553
#1554
#1555
#1556
#1557
#1558
#1559
#1560
#1561
#1562
#1563
#1564
#1565
#1566
#1567
#1568
#1569
#1570
#1571
#1572
#1573
#1573
#1574
#1575
#1576
#1578
#1579
#1580
#1581
#1582
#1583
#1584
#1585
#1586
#1587
#1588
#1589
#1590
#1591
#1592
#1593
#1594
#1595
#1596
#1597
#1598
#1599
#1600
#1601
#1602
#1603
#1604
#1605
#1606
#1607
#1608
#1609
#1610
#1611
#1612
#1613
#1614
#1616
#1617
#1618
#1619
#1620
#1621
#1622
#1623
#1624
#1625
#1626
#1627
#1628
#1629
#1630
#1631
#1632
#1634
#1634
#1635
#1636
#1638
#1639
#1639
#1640
#1641
#1642
#1643
#1644
#1645
#1646
#1646
#1647
#1648
#1649
#1649
#165
#1650
#1651
#1652
#1653
#1654
#1655
#1656
#1659
#166
#1660
#1661
#1662
#1663
#1664
#1669
#1670
#1671
#1672
#1672
#1673
#1674
#1675
#1676
#1677
#1678
#168
#1680
#1682
#1682
#1683
#1683
#1684
#1685
#1686
#1687
#1688
#1689
#1690
#1691
#1692
#1693
#1694
#1695
#1695
#1696
#1697
#1698
#1700
#1701
#1702
#1703
#1703
#1704
#1705
#1706
#1707
#1709
#1709
#1710
#1711
#1712
#1713
#1714
#1715
#1715
#1716
#1717
#1718
#1719
#1719
#1721
#1721
#1722
#1722
#1724
#1725
#1725
#1726
#1727
#1729
#1730
#1731
#1733
#1733
#1734
#1735
#1736
#1737
#1738
#1739
#1740
#1741
#1742
#1743
#1744
#1745
#1746
#1747
#1748
#1749
#1750
#1751
#1753
#1754
#1755
#1756
#1757
#1758
#1759
#1760
#1761
#1763
#1764
#1765
#1766
#1767
#1767
#1769
#1769
#1770
#1771
#1773
#1774
#1774
#1775
#1775
#1776
#1781
#1782
#1783
#1784
#1785
#1788
#1789
#1790
#1791
#1792
#1793
#1794
#1795
#1795
#1796
#1797
#1798
#1799
#1800
#1801
#1802
#1802
#1803
#1804
#1805
#1806
#1807
#1807
#1808
#1809
#1809
#181
#1810
#1810
#1811
#1815
#1816
#1816
#1817
#1818
#1823
#1824
#1825
#1825
#1826
#1827
#1827
#1828
#1829
#1830
#1831
#1832
#1833
#1834
#1835
#1836
#1837
#1838
#1839
#1840
#1841
#1841
#1842
#1843
#1844
#1845
#1846
#1847
#1847
#1849
#1853
#1854
#1855
#1855
#1856
#1857
#1858
#1860
#1860
#1861
#1862
#1863
#1864
#1867
#1868
#1869
#1870
#1872
#1873
#1874
#1877
#1878
#1878
#1879
#188
#1880
#1881
#1881
#1882
#1883
#1883
#1884
#1885
#1885
#1886
#1886
#1887
#1887
#1888
#1889
#1891
#1892
#1893
#1894
#1897
#1898
#1899
#1899
#1900
#1901
#1903
#1904
#1905
#1906
#1907
#1908
#1908
#1909
#1909
#1910
#1910
#1911
#1911
#1912
#1912
#1913
#1914
#1915
#1916
#1917
#1917
#1918
#1918
#1919
#1919
#1920
#1920
#1921
#1922
#1923
#1924
#1925
#1927
#1927
#1928
#1929
#1930
#1931
#1932
#1934
#1935
#1936
#1936
#1937
#1938
#1939
#1941
#1942
#1943
#1944
#1944
#1945
#1946
#1947
#1948
#1949
#1950
#1950
#1951
#1952
#1953
#1953
#1954
#1954
#1955
#1956
#1957
#1957
#1958
#1959
#1959
#196
#1960
#1962
#1963
#1964
#1965
#1966
#1967
#1968
#197
#1970
#1971
#1972
#1972
#1973
#1973
#1975
#1976
#1976
#1977
#1977
#1978
#1978
#1979
#198
#1980
#1981
#1982
#1983
#1984
#1985
#1986
#1986
#1987
#1988
#1988
#1989
#199
#1990
#1990
#1991
#1991
#1992
#1993
#1993
#1994
#1995
#1996
#1997
#1998
#1999
#1999
#2000
#2003
#2004
#2004
#2005
#2006
#2007
#2008
#2009
#2009
#2010
#2012
#2013
#2014
#2017
#2017
#2018
#2019
#202
#2020
#2021
#2022
#2023
#2024
#2025
#2026
#2027
#2028
#2029
#2030
#2031
#2032
#2033
#2034
#2035
#2036
#2037
#2037
#2038
#2039
#2040
#2040
#2041
#2042
#2043
#2044
#2045
#2045
#2046
#2047
#2048
#2049
#2049
#205
#2050
#2051
#2052
#2053
#2054
#2055
#2056
#2057
#2058
#2058
#2059
#2059
#206
#2060
#2061
#2062
#2063
#2064
#2065
#2066
#2067
#2068
#2069
#2069
#2070
#2071
#2072
#2073
#2074
#2075
#2076
#2077
#2077
#2078
#2078
#2079
#208
#2080
#2081
#2082
#2083
#2084
#2085
#2085
#2086
#2087
#2088
#2089
#2089
#209
#2090
#2092
#2093
#2094
#2094
#2096
#2097
#2098
#2099
#210
#2100
#2101
#2102
#2103
#2104
#2105
#2106
#2107
#2108
#2109
#2110
#2111
#2112
#2113
#2114
#2115
#2116
#2117
#2118
#2119
#2119
#2120
#2121
#2122
#2124
#2125
#2126
#2128
#2129
#2129
#2130
#2132
#2133
#2134
#2135
#2136
#2138
#2138
#2139
#214
#2140
#2141
#2142
#2143
#2144
#2145
#2146
#2147
#2148
#2149
#2150
#2151
#2152
#2153
#2154
#2155
#2156
#2157
#2159
#2161
#2161
#2162
#2163
#2164
#2164
#2165
#2166
#2166
#2167
#2167
#2169
#2170
#2171
#2172
#2173
#2173
#2174
#2175
#2176
#2177
#2178
#2179
#2180
#2181
#2182
#2183
#2184
#2185
#2185
#2186
#2186
#2187
#2187
#2188
#219
#2190
#2191
#2192
#2193
#2194
#2195
#2195
#2196
#2197
#2197
#2198
#2198
#2199
#2200
#2201
#2202
#2203
#2205
#2205
#2206
#2207
#2208
#2210
#2211
#2211
#2212
#2213
#2213
#2214
#2215
#2216
#2217
#2218
#2218
#2219
#2221
#2221
#2222
#2223
#2223
#2224
#2225
#2226
#2227
#2228
#223
#2231
#2232
#2234
#2235
#2235
#2236
#2236
#2237
#2237
#2238
#224
#2240
#2240
#2241
#2242
#2243
#2244
#2244
#2245
#2246
#2246
#2247
#2247
#2248
#2248
#2249
#2250
#2250
#2251
#2251
#2252
#2253
#2254
#2255
#2256
#2257
#2258
#2259
#2260
#2261
#2262
#2263
#2264
#2265
#2266
#2266
#2267
#2268
#2269
#227
#2270
#2271
#2272
#2273
#2274
#2275
#2276
#2277
#2278
#2279
#2279
#2280
#2282
#2283
#2283
#2285
#2286
#2287
#2288
#2289
#2290
#2291
#2292
#2292
#2293
#2294
#2295
#2295
#2296
#2298
#2299
#2300
#2301
#2302
#2303
#2304
#2305
#2306
#2307
#2308
#2309
#2310
#2312
#2313
#2313
#2314
#2315
#2315
#2316
#2316
#2317
#2318
#2319
#2320
#2321
#2322
#2323
#2323
#2324
#2325
#2327
#2328
#2329
#2329
#2331
#2331
#2332
#2333
#2334
#2335
#2336
#2338
#2339
#2340
#2340
#2341
#2342
#2343
#2343
#2344
#2345
#2345
#2346
#2347
#2348
#2349
#2351
#2352
#2353
#2354
#2355
#2356
#2357
#2358
#2359
#2360
#2360
#2361
#2361
#2363
#2364
#2365
#2366
#2367
#2367
#2368
#2368
#2369
#2369
#2370
#2370
#2371
#2372
#2372
#2373
#2374
#2375
#2376
#2377
#2378
#2379
#2380
#2381
#2382
#2383
#2384
#2385
#2386
#2387
#2388
#2389
#2390
#2391
#2392
#2393
#2394
#2395
#2396
#2397
#2398
#2399
#2400
#2401
#2401
#2402
#2403
#2404
#2404
#2405
#2406
#2406
#2407
#2408
#2409
#2409
#241
#2410
#2411
#2412
#2413
#2414
#2415
#2415
#2416
#2416
#2417
#2418
#2418
#2419
#2419
#242
#2420
#2420
#2421
#2421
#2422
#2422
#2423
#2425
#2426
#2427
#2428
#2429
#2429
#2430
#2430
#2431
#2431
#2432
#2433
#2434
#2435
#2436
#2437
#2438
#2439
#244
#2440
#2440
#2441
#2441
#2443
#2444
#2445
#2446
#2447
#2448
#2449
#2450
#2451
#2452
#2453
#2453
#2454
#2454
#2455
#2455
#2457
#2458
#2459
#2459
#246
#2460
#2461
#2461
#2462
#2463
#2464
#2465
#2465
#2466
#2467
#2467
#2468
#2469
#2470
#2471
#2472
#2472
#2473
#2473
#2474
#2474
#2475
#2475
#2476
#2476
#2477
#2478
#2479
#248
#2480
#2483
#2484
#2484
#2485
#2485
#249
#2490
#2491
#2492
#2493
#2493
#2494
#2495
#2495
#2496
#2496
#2497
#2497
#2498
#2498
#2499
#2500
#2501
#2502
#2503
#2503
#2504
#2505
#2506
#2507
#2507
#2508
#2508
#2509
#251
#2511
#2512
#2513
#2514
#2515
#2516
#2516
#2517
#2517
#2518
#2519
#2519
#252
#2522
#2522
#2523
#2524
#2525
#2527
#2527
#2528
#2529
#253
#2530
#2531
#2532
#2533
#2534
#2534
#2535
#2535
#2536
#2536
#2537
#2537
#2539
#2539
#254
#2540
#2540
#2541
#2541
#2544
#2545
#2545
#2547
#2548
#2548
#2549
#2550
#2550
#2552
#2552
#2553
#2553
#2554
#2554
#2555
#2555
#2556
#2556
#2557
#2558
#2558
#2559
#2559
#2560
#2560
#2562
#2562
#2564
#2564
#2565
#2566
#2566
#2567
#2567
#2568
#2568
#2569
#2569
#257
#2571
#2572
#2572
#2573
#2574
#2574
#2575
#2575
#2576
#2576
#2577
#2577
#2578
#2578
#2579
#2579
#258
#2580
#2580
#2581
#2582
#2582
#2583
#2583
#2584
#2585
#2585
#2586
#2587
#2587
#2588
#2589
#2589
#2590
#2591
#2592
#2594
#2594
#2595
#2596
#2596
#2597
#2597
#2598
#2598
#2599
#2599
#260
#2600
#2600
#2601
#2601
#2602
#2602
#2603
#2603
#2604
#2604
#2605
#2605
#2606
#2606
#2607
#2607
#2608
#2608
#2609
#2609
#2610
#2610
#2611
#2611
#2612
#2612
#2613
#2613
#2614
#2615
#2616
#2617
#2617
#2618
#2618
#2619
#2620
#2620
#263
#266
#267
#269
#272
#274
#275
#276
#277
#278
#282
#283
#284
#285
#286
#287
#289
#29
#290
#291
#292
#293
#294
#295
#296
#297
#299
#300
#301
#302
#303
#304
#305
#306
#308
#309
#310
#311
#312
#313
#314
#315
#316
#317
#317
#318
#319
#320
#321
#322
#323
#324
#324
#325
#327
#330
#331
#332
#333
#334
#336
#337
#337
#338
#339
#341
#342
#343
#343
#344
#345
#346
#347
#348
#349
#350
#351
#352
#352
#353
#354
#355
#356
#357
#358
#359
#361
#363
#364
#365
#366
#367
#367
#368
#369
#372
#373
#374
#375
#376
#377
#378
#379
#381
#382
#383
#384
#386
#388
#389
#390
#391
#392
#393
#394
#395
#396
#397
#398
#399
#401
#402
#402
#403
#404
#405
#406
#407
#408
#409
#410
#411
#412
#413
#414
#415
#416
#417
#418
#419
#420
#421
#422
#423
#424
#425
#426
#427
#427
#429
#430
#431
#432
#433
#434
#435
#436
#437
#440
#441
#442
#443
#444
#445
#446
#447
#448
#449
#450
#451
#452
#452
#453
#455
#456
#457
#459
#460
#461
#462
#463
#464
#465
#466
#467
#469
#470
#473
#474
#475
#476
#477
#478
#479
#480
#481
#482
#483
#484
#486
#486
#487
#489
#490
#492
#493
#494
#495
#496
#497
#498
#499
#502
#504
#505
#507
#508
#509
#510
#511
#513
#514
#515
#516
#517
#519
#520
#521
#521
#522
#523
#524
#525
#526
#527
#529
#530
#531
#532
#533
#534
#535
#536
#538
#539
#540
#541
#542
#544
#545
#546
#547
#548
#550
#551
#552
#553
#554
#555
#556
#557
#558
#561
#562
#563
#564
#564
#566
#567
#571
#573
#574
#575
#576
#577
#578
#579
#580
#582
#583
#584
#585
#586
#587
#588
#589
#590
#590
#591
#592
#594
#596
#598
#599
#605
#606
#607
#608
#609
#610
#611
#612
#613
#615
#615
#616
#617
#618
#620
#622
#623
#624
#624
#625
#626
#627
#629
#630
#631
#632
#633
#634
#635
#636
#637
#638
#639
#640
#641
#642
#643
#644
#645
#646
#648
#649
#649
#650
#651
#652
#653
#653
#654
#655
#655
#656
#657
#659
#660
#661
#663
#664
#665
#666
#667
#668
#669
#670
#671
#672
#673
#674
#675
#676
#677
#678
#679
#680
#681
#682
#683
#684
#685
#686
#687
#689
#69
#690
#691
#692
#693
#694
#695
#697
#698
#699
#700
#701
#702
#703
#704
#705
#706
#707
#708
#709
#710
#711
#714
#715
#715
#717
#718
#719
#720
#721
#722
#723
#723
#726
#727
#728
#730
#730
#731
#734
#735
#736
#737
#738
#740
#742
#743
#743
#744
#745
#746
#747
#748
#749
#752
#752
#753
#754
#755
#756
#757
#758
#759
#760
#762
#764
#765
#767
#768
#769
#77
#772
#773
#774
#776
#777
#778
#779
#78
#780
#781
#782
#782
#783
#784
#785
#786
#787
#788
#789
#79
#790
#792
#793
#794
#798
#800
#801
#802
#803
#804
#805
#807
#807
#808
#809
#810
#811
#812
#813
#814
#815
#816
#816
#817
#818
#819
#820
#822
#823
#824
#825
#826
#826
#827
#829
#830
#831
#832
#832
#833
#834
#835
#835
#836
#836
#837
#838
#839
#840
#840
#841
#842
#843
#844
#845
#846
#849
#851
#852
#853
#854
#855
#856
#858
#859
#860
#861
#862
#863
#864
#864
#865
#866
#867
#870
#870
#872
#872
#873
#875
#876
#877
#878
#879
#880
#880
#881
#882
#883
#884
#885
#886
#887
#888
#889
#890
#891
#892
#894
#895
#896
#896
#897
#898
#899
#900
#901
#902
#903
#904
#905
#905
#906
#907
#908
#908
#910
#911
#912
#913
#914
#915
#916
#917
#918
#919
#919
#920
#921
#921
#922
#923
#924
#925
#926
#927
#928
#929
#930
#930
#931
#931
#933
#934
#935
#936
#937
#938
#939
#940
#941
#941
#942
#943
#944
#945
#946
#947
#948
#948
#950
#951
#952
#953
#954
#955
#956
#958
#961
#962
#963
#963
#964
#965
#966
#967
#968
#969
#970
#971
#972
#973
#974
#975
#976
#976
#977
#978
#979
#979
#980
#982
#983
#984
#985
#986
#986
#987
#988
#989
#990
#991
#992
#993
#994
#995
#996
#997
#998
#999
0.23.0
0.23.1
0.23.2
0.25.0
0.27.0
0.27.1
0.29.0
0.31.0
0.33.0
0.33.1
0.35.0
0.35.1
0.37.0
0.39.0
0.40.0
0.41.0
0.41.0rc1
0.41.0rc2
0.41.0rc3
0.42.0
0.42.0dev
0.42.0rc0
0.42.0rc1
0.42.0rc2
0.43.0
0.43.0rc0
0.43.0rc1
0.43.0rc2
0.43.0rc3
0.43.0rc4
0.44.0
0.44.0dev
0.44.0rc0
0.44.0rc1
0.44.0rc2
0.44.0rc3
0.44.0rc4
0.44.0rc5
0.45.0
0.45.0dev
0.45.0rc0
0.45.0rc1
0.45.0rc2
0.45.0rc3
0.46.0
0.46.0dev
0.46.0rc0
0.46.0rc1
0.46.0rc2
0.46.1
0.47.0
0.47.0dev
0.47.0rc0
0.47.0rc1
0.47.0rc2
0.48.0dev
Select branches
Hide Pull Requests
1679-add-mbridge-dsitillation-for-puzzletron
Model-Optimizer-Agent
aanoosheh/topk-kd-topp-ghost
add-codex-claude-skill-wrappers
agent/fix-aa-next-openhands-timeout
agent/fix-compressed-kimi-ptq
agent/kv-cache-autoquant-pr
agent/pin-nel-launcher-version
ajrasane/diffusers_cudagraphs
ajrasane/diffusion_autodeploy
ajrasane/dynamo-onnx-export
ajrasane/dynamo_export
ajrasane/efficient_vit
ajrasane/normalize-calibrated-request-contract
ajrasane/nvbug-5612316-fp8-attention-qdq
ajrasane/nvbug-6721556-swin-mixed-dtype-add
ajrasane/nvbug_6034518
ajrasane/ort-patching-capability-split
ajrasane/qwen3-omni-30B
ajrasane/qwen3omni_final
ajrasane/readme_update
ajrasane/reduce-diffusers-test-compute
ajrasane/remove-ensure-default-opset
ajrasane/sparse-quant-trt-example
akuriparambi/fix-aq-score-scaling-20260623
alit/moe_dev
asma/MoE_amax_sync
asma/activation-quant-aware-scale-setting
asma/add_support_for_custom_backend
asma/aq_bug_fix
asma/autoquantize-qkv-self-attn-scoring
asma/beebot-speed-benchmark-20260706
asma/distributed-amax-error-handling
asma/fix-nvbug-6532161-fsdp2-tied-params
asma/fix-nvfp4-multi-gpu-export
asma/fp4-stochastic-rounding
asma/fp8-sweep-offset-range
asma/fsdp2_full_state
asma/gptq-scale-algorithms
asma/laq-algorithm
asma/laq-export-support
asma/laq-int3-int2-scale-learning
asma/layerwise_skip_in_meta
asma/mcore-aq
asma/mcore_mse_fp32
asma/nemotron_aq
asma/quant_backends_initial_support
asma/qwen3.8_fp4
asma/qwen38-day0-local-hessian-recipes
asma/scale_learning
asma/sequential-calibrate
asma/super_nvfp4_recipe
asma/tensorquantizer-random-hadamard
asma/vlm-layerwise-calibration
asma/wmse
auto/bump-uv-lock-main-2026-09-28
autocast_resize_fp16_support
backup/24-sparsity-pre-rebuild
bchislett/develop_stash
bchislett/eagle-optimize-dynamic-seqlen
bchislett/eagle-tensorboard-logs
bchislett/eagle-training-qol-moe-etc
bchislett/gpt-oss-120b-EAGLE3-round5
bchislett/specbench-tweaks
bchislett/specdec_bench_vllm_flags
beebot/nvbug-6311597
beebot/nvbug-6359821
chad/eval-model-card-provenance
chenhany/OMNIML-5091-cell-t0-d7
chenhany/autonomous-runner-carve-out
chenhany/dataset-scripts-and-specdec-readme
chenhany/dflash
chenhany/dflash-deepseek-v4-flash
chenhany/eagle3-qwen06b
chenhany/fix-dflash-yaml
chenhany/fix-nvrx-compat
chenhany/gemma-4-vllm-image-fix
chenhany/mmlu-moe-ep-shard-fix
chenhany/omniml-5352-ptyche-mcp
chenhany/omniml-5438-mcp-mlflow
chenhany/qad-extra-state-fix
chenhany/specdec-bench-phase-1a-provenance-and-s3-upload
chenhany/vlm_sft_dataset
chenhany@nvidia.com/omniml-5492
chenjiel/demo_mxfp4_nvfp4
chenjiel/dsv4_cast
chenjiel/eval-skill-tau3-banking
chenjiel/fastmegatron_fp8
chenjiel/fix_moe_calib
chenjiel/fix_seq
chenjiel/int4
chenjiel/iq-1bit-2bit-formats
chenjiel/iq-export-reuse-cuda-decode
chenjiel/iq-kernel-dedup
chenjiel/moe
chenjiel/nvfp4-activation-calib-study
chenjiel/nvfp4-global-scale-study
chenjiel/qwen3-8b-online-eagle3
chenjiel/recipe-nvfp4-experts-mse-fp8-cast-kv
chenjiel/recipe-nvfp4-experts-mse-fp8-cast-kv-2
chenjiel/refactor_qwen35
chenjiel/trim-hf-ptq-example-tests
cjluo-nv-patch-1
cjluo-nv-patch-2
coderabbitai/chat/ea0c193
coderabbitai/docstrings/93bab27
coderabbitai/utg/93bab27
codex/megatron-embedding-weight-quantization
codex/qwen36-fp8-granularity-study
codex/secure-slurm-env-setup
depth_importance_example
dkorzekwa/anymodel_layer_name_predicates
dkorzekwa/claude_qwen35
dkorzekwa/claude_qwen35_distill
dkorzekwa/claude_qwen35_eval
dkorzekwa/doge_datablend_weights_tuning
dkorzekwa/hf_qad_example_clarity
dkorzekwa/kd-gap-data-blend-tuning
dkorzekwa/puzzletron_claude_skill
dkorzekwa/qad_docs
dla_surgeon_integration
docs/eval-mlflow-verification
docs/ptq-pytorch-base-image
edge_llm_example
eval-skill-multiarch-cu130
export_dedup_bkstp
export_dedup_cjluo
export_dedup_cjluo2
export_dedup_simplify
export_dedup_verify
feat/mbridge-qad-pipeline-fixes
feat/nvfp4-conv3d-export
feat/scoped-calibration-algo-cfg
feature/codex-day0-agent-roles
feature/dasc-state-sparsity
feature/dasc-state-sparsity-complete-layers
feature/dasc-state-sparsity-explicit-sort
feature/dasc-state-sparsity-identity-restore
feature/dasc-state-sparsity-lifecycle-fixes
feature/dasc-state-sparsity-parity-gate
feature/dasc-state-sparsity-review-api
feature/dasc-state-sparsity-review-boundary
feature/dasc-state-sparsity-review-contract
feature/dasc-state-sparsity-review-contract-cleanup
feature/dasc-state-sparsity-review-device-independent
feature/dasc-state-sparsity-review-device-test
feature/dasc-state-sparsity-review-final
feature/dasc-state-sparsity-review-final2
feature/dasc-state-sparsity-review-fixes
feature/dasc-state-sparsity-review-policy-device
feature/dasc-state-sparsity-review-public-api
feature/dasc-state-sparsity-review-real-scalars
feature/dasc-state-sparsity-review-rounding
feature/dasc-state-sparsity-review-rounding-cleanup
feature/dasc-state-sparsity-review-validation
feature/dasc-state-sparsity-review-validation-source
feature/dasc-state-sparsity-storage-contract
feature/dasc-state-sparsity-storage-roundtrip
feature/dasc-state-sparsity-validation-helpers
feature/mcore_mse_mixed_precision
feature/puzzletron_v2
feature/qwen-image-dmd2
feature/qwen-image-dmd2-qat
feature/qwen-image-fastgen-qad
feature/qwen-image-svdquant-nvfp4
feature/runtime_opt_vllm
fix-nemotronh-fused-awq-export
fix-nvfp4-gptq-static-export
fix/dynamo-onnx-export
fix/gemma4-vocab-size-export
fix/hf-ptq-pad-token-overwrite
fix/megatron-bridge-untied-lm-head-quantization
fix/modelopt-mcp-log-fetch
fix/nixl-efa-libfabric
fix/nvfp4-static-global-amax-tp-ep-sync
fix/puzzletron-campaign-reliability
fix/specdec-bench-provenance
fix_auto_cast_io_special_nodes_bug
fridah/4_6_act
fridah/algo-cfg-1-capabilities
fridah/algo-cfg-2-write-mask
fridah/algo-cfg-3-compile
fridah/algo-cfg-4-execute
fridah/blog-single-gpu-ptq
fridah/export-module-split
fridah/fidelity-check
fridah/fix-fp8-sweep
fridah/four-over-six-algorithm
fridah/general-fp8-sweep
fridah/glm-tmp
fridah/glm5.1-tmp
fridah/global-hessian
fridah/global-hessian-cw
fridah/improve-local-hessian
fridah/in_exclude_mod
fridah/k3-layerwise-fused-export
fridah/k3-layerwise-fused-export-prerebase
fridah/layerwise-kv-cache-replay-fix
fridah/layerwise-mse
fridah/layerwise-mse-fix
fridah/layerwise-mtp-support
fridah/local-hessian-export
fridah/moe-gate-up-sync-cleanup
fridah/mse-dss
fridah/quantile-calibrator
fridah/super-awq
gh-pages
gkarch/add_attn_pruning
gkarch/add_lmms_eval
gkarch/add_mamba_support_to_runtime_stats
gkarch/add_moe_to_runtime_stats
gkarch/multi-gpu-vllm-benchmarking
gkarch/nemotron-nano-v3
gkarch/puzzletron_v2-nano-dockerfile
gkarch/puzzletron_v2_llama
gkarch/puzzletron_v2_sync_20260730
gkarch/support_nemotronh
gkarch/sync-main-449a3992
gkarch/test-image-build
haoguo/dflash-cosmos3-launcher-example
haoguo/dflash-finetune-example
haoguo/dflash-fp32-master-weights
haoguo/dflash-rope-fp32
haoguo/dflash-training-perf
haoguo/dsv3-eagle
haoguo/eagle-doc
haoguo/eagle-newtrainer
haoguo/fix-0208
haoguo/fix-dflash-rope-theta-rope-parameters
haoguo/fix-launcher-global-vars
haoguo/fix-restore
haoguo/fixci-0611
haoguo/gemma4-e4b-support
haoguo/hardware_autoquant
haoguo/k25dflash
haoguo/launcher-claudemd-scope
haoguo/m3-dspark
haoguo/mla-eagle-K2
haoguo/model_patches
haoguo/nvbug6304444
haoguo/per-model-recipes
haoguo/qwen3-8b-dspark-online-example
haoguo/qwen3-8b-dspark-streaming-example
haoguo/qwen3vl
haoguo/qwen3vl-mrope
haoguo/sgl-backend
haoguo/skierat-cr2-guide
haoguo/spec-doc
haoguo/spec-mixin
haoguo/spec-mixin-new
haoguo/spec-recipe-import
haoguo/streaming-vllm029-fixes
haoguo/temp-offline-eagle-ci
haoguo/test-offline-ci
haoguo/udpate-dflash-recipe
haoguo/vllm-dumper
haoxiz/autocast-mismatch
haoxiz/env-fix
haoxiz/kv_cache
haoxiz/onnx-ptq-trt-abi
haoxiz/onnx-validation
haoxiz/trt-fix
hthadicherla/add-ort-preload-dll-cudnn
hthadicherla/torch-ptq-onnx
hungyuehc/omniml-4944
hungyuehc/omniml-5051
hungyuehc/omniml-5395
hungyuehc/omniml-5415
hungyuehc/omniml-5823-iq2xs-nemotron-recipe
hungyuehc/omniml-5899-cuda
hungyuehc/omniml-5899-export
hungyuehc/omniml-5899-ggml
hungyuehc/omniml-5899-tests-v2
hungyuehc/q8-0-export-recipes
hungyuehc/q8-0-quantization
iputterman/specdec-04-15
iq2xs-unified-export
jennifchen/config_compat
jennifchen/fix_kv_resume
jennifchen/mcore_autoquant
jennifchen/moe_lora
jennifchen/moe_lora_with_sequentialmlp
jennifchen/super_nvfp4_recipe
jennifchen/te_qad_debug
jingyux/24-sparsity
jingyux/24-sparsity-backup-20260708
jingyux/24-sparsity-test
jingyux/3-4-diffusion
jingyux/block-quant-onnx
jingyux/diffusion-autoquant
jingyux/diffusion-skip-softmax
jingyux/diffusion.export
jingyux/fastvideo-0
jingyux/fixed-ltx2-ootm
jingyux/headsparsity
jingyux/kimi-k2
jingyux/kimi-k2-nvfp4
jingyux/lobcq
jingyux/modelopt-peft-te
jingyux/qwen-image-dmd2-svdquant-peft
jingyux/qwen-image-fastgen-qad
jingyux/skipsoftmax-diffusion
jingyux/svd-export
jingyux/svd-mcore
jingyux/triton_fa_skip_softmax_diffusion
jingyux/triton_fa_skip_softmax_diffusion_exp
jingyux/vllm-calib
jingyux/vsa-diffusion
jingyux/vsa-training
jingyux/wan22-svdquant-export
jrausch/anymodel_lmeval
jrausch/pr1769-review
jrausch/puzzletron-gpu-ci-image-workflow
jrausch/puzzletron-gpu-quality-baseline-v4
jrausch/puzzletron-v2-ci-quality-closeout
jrausch/puzzletron-v2-config-simplification
jrausch/puzzletron-v2-qwen-runtime-equivalence
jrausch/puzzletron-v2-structured-results-reporting
jrausch/qwen35-4b-vlm-competitive-campaign-followup-pr2361
juhi/nemotron_35
juhim/autoquant-recipe
juhim/haq_poc
kaix/2_4_sa
kaix/agent-marketplace
kaix/dasc-state-sparsity
kaix/kaix/sparse_attention_calibration_decode
kaix/linear-attention-ci-cleanup
kaix/linear-attention-decode-first
kaix/linear-attention-decode-triton
kaix/linear-attention-qat-m1
kaix/linear-attention-qat-m2
kaix/linear-attention-qat-m3
kaix/linear-attention-qat-m4
kaix/linear-attention-qat-m5
kaix/linear-attention-vllm
kaix/mask-reuse-01-offline-calibration
kaix/mask-reuse-02-vllm-capture
kaix/mask-reuse-03-topology-selection
kaix/mask-reuse-calib
kaix/mla_opt
kaix/modelopt_agent
kaix/skip_softmax_calib
kaix/skip_softmax_calib_formula
kaix/skip_softmax_triton
kaix/sparse_attention_export
kaix/sparse_attention_llm_eval
kaix/sparse_attention_with_triton_backend
kaix/sparse_attn_integration
kaix/sparse_attn_quant
kaix/sparse_attn_quant_mixed_fp16
kaix/sparse_attn_quant_v2
kaix/splitk_skip_softmax
kaix/triattn
kaix/triton-skip-softmax-calib
kaix/triton_decode_skip_softmax
kaix/vllm_decode_skip_softmax
kaix/vllm_skip_calib
kaix/watersic-kv-cache
kinjal/automodel
kinjal/fix_megatron_vllm_fq_export
kinjal/fix_vllm_env_var
kinjal/fix_vllm_fq
kinjal/grouped_tp
kinjal/latest_vllm
kinjal/skip_kv_tokens
kinjal/sungsooh_parallel-fakequant-export-pr
kinjal/vllm_app
kinjal/vllm_kvquant_n
kinjal/vllm_wrapper
kmorabia/mbridge-glm53-flash
kmorabia/prune-gemma4-e4b
kmorabia/scicode-verbosity-mitigation
lmikaelyan/compress-vlms
lmikaelyan/fix-generation-params-pruning
main
mchochowski/puzzle_v2_qwen3.5_0p8
mchochowski/puzzletron_distillation
mchochowski/test-image-build
mxin/fix-export-megatron
mxin/fp4-new-kernel
mxin/general-real-quant-refit
mxin/kv-boundary-skip
mxin/ltx-distill-v2
mxin/modelopt-real-quant-canonical
mxin/modelopt-real-quant-refactor
mxin/nvfp4-kv-scale1-minimal
mxin/pdd-fastgen-refactor
mxin/recipe-defaults
mxin/rl-nano3
mxin/sglang
mxin/test-pr
mxin/vllm_fix
mxin/yaml-include
northbloom
oncallbot/fix-795
pensieve-intern/OMNIML-4634/hidden-state-dump-support
pensieve-intern/OMNIML-4646/hidden-state-dump-support
pensieve-intern/OMNIML-4646/training-support
pensieve-intern/OMNIML-4681/hidden-state-dump-support
pensieve-intern/OMNIML-4681/synth-support
pensieve-intern/OMNIML-4681/training-support
pensieve-intern/OMNIML-4691/hidden-state-dump-support
pensieve-intern/OMNIML-4691/synth-support
pensieve-intern/OMNIML-4691/training-support
pensieve-intern/OMNIML-4701/synth-support
pensieve-intern/OMNIML-4735/deployment-support
pensieve-intern/OMNIML-4745/synth-support
pensieve-intern/OMNIML-4755/synth-support
pensieve-intern/OMNIML-4872/author-yaml
pensieve-intern/OMNIML-4879/cell-t0-d3
pensieve-intern/OMNIML-4879/cell-t0-d7
pensieve-intern/OMNIML-4879/cell-t1-d3
pensieve-intern/OMNIML-4879/cell-t1-d7
pensieve-intern/OMNIML-4885/cell-t0-d3
pensieve-intern/OMNIML-4885/cell-t0-d7
pensieve-intern/OMNIML-4885/cell-t1-d3
pensieve-intern/OMNIML-4885/cell-t1-d7
pensieve-intern/OMNIML-4885/t0_d3
pensieve-intern/OMNIML-4886/t0_d3
pensieve-intern/OMNIML-4926/cell-t0-d3
pensieve-intern/OMNIML-4926/cell-t0-d7
pensieve-intern/OMNIML-4926/cell-t1-d3
pensieve-intern/OMNIML-4927/t0_d3
pensieve-intern/OMNIML-4928/t0_d7
pensieve-intern/OMNIML-4929/t1_d3
pensieve-intern/OMNIML-4930/t1_d7
pensieve-intern/OMNIML-4961/cell-t0-d7
pensieve-intern/OMNIML-4961/cell-t1-d3
pensieve-intern/OMNIML-4961/cell-t1-d7
pensieve-intern/OMNIML-4962/t0_d3
pensieve-intern/OMNIML-4967/cell-t0-d3
pensieve-intern/OMNIML-4967/cell-t0-d7
pensieve-intern/OMNIML-4967/cell-t1-d3
pensieve-intern/OMNIML-4967/cell-t1-d7
pensieve-intern/OMNIML-5022/cell-t0-d3
pensieve-intern/OMNIML-5022/cell-t1-d7
pensieve-intern/OMNIML-5022/t0_d7
pensieve-intern/OMNIML-5057/cell-t0-d7
pensieve-intern/OMNIML-5081/cell-t0-d7
pensieve-intern/OMNIML-5088/cell-t0-d7
pensieve-intern/OMNIML-5095/cell-t0-d3
pensieve-intern/OMNIML-5095/cell-t0-d7
pensieve-intern/OMNIML-5095/cell-t1-d3
pensieve-intern/OMNIML-5095/cell-t1-d7
pensieve/fix-issue-1041
pensieve/fix-issue-1066
pensieve/fix-issue-5997673
pensieve/fix-issue-5997787
pensieve/fix-issue-5997832
pinedrift/dflash2-capture-fix
pull-request/1016
pull-request/1017
pull-request/1027
pull-request/1037
pull-request/1043
pull-request/1105
pull-request/1118
pull-request/1152
pull-request/1153
pull-request/1171
pull-request/1190
pull-request/1213
pull-request/1214
pull-request/1241
pull-request/1272
pull-request/1297
pull-request/1311
pull-request/1315
pull-request/1333
pull-request/1344
pull-request/1378
pull-request/1391
pull-request/1398
pull-request/1415
pull-request/1431
pull-request/1499
pull-request/1504
pull-request/1507
pull-request/1523
pull-request/1530
pull-request/1541
pull-request/1649
pull-request/1725
pull-request/1767
pull-request/1769
pull-request/1774
pull-request/1775
pull-request/1784
pull-request/1802
pull-request/1807
pull-request/1809
pull-request/1816
pull-request/1825
pull-request/1886
pull-request/1944
pull-request/1986
pull-request/1999
pull-request/2040
pull-request/2077
pull-request/2078
pull-request/2094
pull-request/2138
pull-request/2166
pull-request/2167
pull-request/2195
pull-request/2198
pull-request/2218
pull-request/2223
pull-request/2240
pull-request/2247
pull-request/2248
pull-request/2251
pull-request/2292
pull-request/2313
pull-request/2315
pull-request/2316
pull-request/2329
pull-request/2360
pull-request/2361
pull-request/2372
pull-request/2415
pull-request/2416
pull-request/2418
pull-request/2419
pull-request/2420
pull-request/2425
pull-request/2430
pull-request/2431
pull-request/2440
pull-request/2441
pull-request/2453
pull-request/2455
pull-request/2459
pull-request/2461
pull-request/2465
pull-request/2467
pull-request/2472
pull-request/2476
pull-request/2496
pull-request/2497
pull-request/2508
pull-request/2516
pull-request/2517
pull-request/2534
pull-request/2552
pull-request/2568
pull-request/2569
pull-request/2576
pull-request/2581
pull-request/2582
pull-request/2587
pull-request/2589
pull-request/2594
pull-request/2596
pull-request/2597
pull-request/2602
pull-request/2604
pull-request/2605
pull-request/2606
pull-request/2615
pull-request/2620
pull-request/324
pull-request/402
pull-request/452
pull-request/564
pull-request/590
pull-request/624
pull-request/649
pull-request/6529944-dynamic-nvfp4-trt-compat
pull-request/653
pull-request/655
pull-request/723
pull-request/743
pull-request/752
pull-request/835
pull-request/840
pull-request/864
pull-request/921
pull-request/930
pull-request/948
pull-request/976
pull-request/979
puzletron_v2_dockerfile
puzzletron/mbridge
puzzletron_v2
puzzletron_v2_orch
qad/te-per-expert-1550
quant_rotate
quant_rotate_models
release/0.27.0
release/0.27.0-windows
release/0.29.0
release/0.31.0
release/0.33.0
release/0.33.0-windows
release/0.35.0
release/0.37.0
release/0.39.0
release/0.40.0
release/0.41.0
release/0.42.0
release/0.43.0
release/0.44.0
release/0.45.0
release/0.46.0
release/0.47.0
rislam/cask-output-quant
rohjoshi/14sparsity-exp
rohjoshi/fast-compressed-tensors-load
rohjoshi/scale-learning-blog
rohjoshi/sparse-kernel
rohjoshi/sparse24-plus-skipsoftmax
shengliangx/config_yaml
shengliangx/deprecating-trtllm-model_type
shengliangx/distributed-export
shengliangx/docs-update
shengliangx/error-old-quant-cfg
shengliangx/hf-ptq-recipe-override
shengliangx/model-download-1-hf-checkpoint-utils
shengliangx/model-download-2-hf-exporters
shengliangx/model-download-3-hf-ptq
shengliangx/model-download-4-megatron
shengliangx/model-download-5-llm-sparsity
shengliangx/move-modeling
shengliangx/move-modeling_1
shengliangx/move-modeling_2
shengliangx/move-modeling_3
shengliangx/move-modeling_4
shengliangx/move-modeling_5
shengliangx/nv-modelopt-cli
shengliangx/prebuild-onnx-ext-off-test-clock
shengliangx/pyrefly-tool
shengliangx/quant_grouped_mm
shengliangx/qwen3vlmoe-export
skierat/dflash-cosmos3-launcher-example
skip-softmax-visualgen-export
ssalian/thinkact
ssameni/puzzletron-bypass
svelury/fp8pb-mxfp8-export
svelury/gptq-data-gettr
svelury/gptq-export
svelury/gptq-lite-helper
svelury/gptq-local-hessian
svelury/gptq-vq-f
svelury/save-last-2-layer-inputs
svelury/seq-calib-save-restore
talora/compute_hidden_states_trtllm_fix
talora/eagle3-experiments
talora/specdec-bench-july-26
temp/day0-harbor-all-fixes-20260918
temp/day0-harbor-five-fixes-20260921
temp/day0-harbor-five-fixes-20260923
template/new-nas-mode
weimingc/autoq-active-moe-nvfp4-mse-exp
weimingc/autoquant-group-scoring
weimingc/autoquant-mtp-recipes
weimingc/autoquant-per-element-group-scoring
weimingc/autoquant-per-element-parent-scoring
weimingc/autoquant-per-element-scoring
weimingc/autoquant_qwen36_lcb_study
weimingc/fix-export-names
weimingc/fix_awq_quant
weimingc/fix_int4_export
weimingc/fix_mcore_nvpf4_export_vllm
weimingc/inkling-attn-support
weimingc/mtp-quantization-export
weimingc/omniml-5271
weimingc/support_mcore_exclude_modules
weimingc/vllm_MoEAWQ
weimingc/vllm_fake_quant_for_fp8_attn_math
yeyu/SWA
yeyu/afterany-pipeline-fix
yeyu/auto_regression_parallel_draft
yeyu/debug
yeyu/debug_paralllel_draft
yeyu/dflash-minimax-m3
yeyu/eagle_embedding_optional
yeyu/eagle_medusa
yeyu/fix_bug_in_loss_mask
yeyu/fix_realquant_bug
yeyu/kv_cache
yeyu/medusa_share_lm_head
yeyu/minimax-m2.7-streaming-dflash
yeyu/mtp-export-rules
yeyu/parallel_eagle
yeyu/pard
yeyu/refactor_multi_step_attn_mask_for_arbitrary_step
yeyu/remove_embedding_from_eagle
yeyu/sage-attention-diffusion
yeyu/set_block
yeyu/skipsoftmax-diffusion
yeyu/specdec-e2e-reference-run
yeyu/specdec-skill-sheets-dflash2-lilicorr
yeyu/tidar_quant
yueshen/export-fused-layernorm-te-spec
yueshen/fp8_qkvo-nvfp4_mlp
yueshen/gemma-auto-quant
yueshen/launcher-megatron-bridge
yueshen/megatron_generate_vlm
yueshen/qad-megatron-bridge
zhiyu/agentic-workflows-schemas
zhiyu/ct-int4-to-gptq-int4
zhiyu/fix-config-export-issue-rope-scaling-type
zhiyu/handle-glm-mtp
zhiyu/patch-dsv3-config
#1
#100
#1000
#1001
#1005
#1006
#1007
#1008
#1009
#1010
#1011
#1012
#1013
#1013
#1014
#1015
#1015
#1016
#1016
#1017
#1017
#1018
#1019
#1020
#1021
#1022
#1023
#1024
#1025
#1026
#1027
#1027
#1028
#1029
#1030
#1031
#1033
#1034
#1035
#1036
#1037
#1037
#1038
#1039
#1040
#1042
#1043
#1043
#1044
#1046
#1047
#1048
#1049
#1050
#1050
#1051
#1052
#1053
#1054
#1054
#1056
#1057
#1058
#1059
#1060
#1061
#1062
#1063
#1065
#1067
#1068
#1069
#1070
#1071
#1072
#1073
#1074
#1075
#1076
#1076
#1077
#1078
#1079
#1080
#1081
#1082
#1082
#1083
#1084
#1085
#1086
#1087
#1089
#1091
#1094
#1096
#1097
#1098
#1099
#1100
#1101
#1102
#1103
#1104
#1105
#1105
#1106
#1107
#1108
#1109
#1110
#1111
#1112
#1113
#1114
#1115
#1116
#1117
#1118
#1118
#1119
#1120
#1121
#1122
#1123
#1124
#1125
#1125
#1126
#1127
#1128
#1129
#1129
#1130
#1131
#1132
#1133
#1134
#1135
#1136
#1137
#1138
#1139
#1140
#1141
#1142
#1143
#1144
#1145
#1146
#1148
#1149
#1150
#1151
#1152
#1152
#1153
#1153
#1154
#1155
#1156
#1157
#1157
#1158
#1159
#1160
#1161
#1162
#1163
#1164
#1165
#1166
#1167
#1168
#1169
#1170
#1171
#1171
#1172
#1174
#1175
#1176
#1177
#1177
#1178
#1179
#1180
#1181
#1182
#1183
#1184
#1185
#1187
#1188
#1189
#1190
#1190
#1191
#1192
#1193
#1194
#1194
#1195
#1196
#1197
#1198
#1199
#1200
#1201
#1202
#1203
#1204
#1205
#1206
#1207
#1208
#1209
#1210
#1211
#1213
#1213
#1214
#1214
#1215
#1216
#1217
#1217
#1218
#1219
#1220
#1221
#1222
#1223
#1224
#1225
#1226
#1227
#1228
#1229
#1230
#1230
#1231
#1232
#1233
#1234
#1235
#1236
#1238
#1239
#1240
#1241
#1241
#1242
#1243
#1244
#1245
#1246
#1247
#1248
#1249
#125
#1250
#1251
#1252
#1253
#1254
#1256
#1257
#1258
#1259
#1260
#1261
#1262
#1263
#1264
#1264
#1265
#1266
#1267
#1268
#1269
#127
#1270
#1271
#1272
#1272
#1273
#1274
#1275
#1276
#1277
#1278
#1279
#128
#1280
#1281
#1284
#1284
#1285
#1286
#1287
#1287
#1288
#1289
#129
#1290
#1291
#1292
#1293
#1295
#1296
#1297
#1297
#1298
#1299
#1300
#1301
#1302
#1303
#1304
#1305
#1306
#1307
#1309
#131
#1310
#1311
#1312
#1313
#1314
#1315
#1315
#1316
#1317
#1318
#1320
#1321
#1322
#1323
#1323
#1324
#1325
#1326
#1327
#1328
#1329
#1330
#1331
#1332
#1333
#1333
#1334
#1335
#1337
#1338
#1338
#1339
#1340
#1341
#1342
#1343
#1344
#1344
#1345
#1347
#1348
#1349
#1350
#1351
#1352
#1353
#1354
#1355
#1356
#1357
#1358
#1359
#1360
#1361
#1362
#1363
#1364
#1364
#1365
#1367
#1367
#1368
#1369
#137
#1370
#1371
#1372
#1373
#1374
#1375
#1376
#1377
#1378
#1378
#1379
#1380
#1381
#1382
#1383
#1384
#1385
#1386
#1387
#1388
#1389
#1390
#1391
#1391
#1392
#1393
#1394
#1395
#1397
#1398
#1398
#1399
#1400
#1401
#1402
#1403
#1404
#1405
#1406
#1406
#1407
#1408
#1409
#1410
#1411
#1412
#1412
#1413
#1413
#1414
#1415
#1415
#1416
#1417
#1418
#1419
#1420
#1421
#1422
#1423
#1425
#1426
#1427
#1428
#1429
#1430
#1431
#1431
#1432
#1433
#1434
#1435
#1436
#1437
#1438
#1439
#1440
#1441
#1442
#1443
#1444
#1445
#1446
#1447
#1448
#1449
#1450
#1451
#1452
#1453
#1454
#1455
#1456
#1457
#1458
#1459
#1460
#1461
#1462
#1463
#1464
#1465
#1465
#1466
#1467
#1467
#1468
#1469
#1470
#1471
#1472
#1473
#1474
#1474
#1475
#1475
#1476
#1477
#1478
#1479
#1480
#1481
#1482
#1483
#1484
#1485
#1486
#1487
#1488
#1489
#1490
#1491
#1491
#1492
#1493
#1494
#1495
#1496
#1497
#1498
#1499
#1499
#1500
#1501
#1502
#1503
#1504
#1504
#1505
#1505
#1506
#1507
#1507
#1508
#1509
#1511
#1512
#1513
#1514
#1515
#1516
#1517
#1518
#1519
#1520
#1520
#1521
#1522
#1523
#1523
#1524
#1525
#1526
#1526
#1527
#1528
#1529
#1530
#1530
#1531
#1532
#1535
#1536
#1537
#1538
#1538
#1540
#1540
#1541
#1541
#1542
#1543
#1544
#1545
#1545
#1546
#1547
#1548
#1549
#1549
#155
#1550
#1551
#1551
#1553
#1553
#1554
#1555
#1556
#1557
#1558
#1559
#1560
#1561
#1562
#1563
#1564
#1565
#1566
#1567
#1568
#1569
#1570
#1571
#1572
#1573
#1573
#1574
#1575
#1576
#1578
#1579
#1580
#1581
#1582
#1583
#1584
#1585
#1586
#1587
#1588
#1589
#1590
#1591
#1592
#1593
#1594
#1595
#1596
#1597
#1598
#1599
#1600
#1601
#1602
#1603
#1604
#1605
#1606
#1607
#1608
#1609
#1610
#1611
#1612
#1613
#1614
#1616
#1617
#1618
#1619
#1620
#1621
#1622
#1623
#1624
#1625
#1626
#1627
#1628
#1629
#1630
#1631
#1632
#1634
#1634
#1635
#1636
#1638
#1639
#1639
#1640
#1641
#1642
#1643
#1644
#1645
#1646
#1646
#1647
#1648
#1649
#1649
#165
#1650
#1651
#1652
#1653
#1654
#1655
#1656
#1659
#166
#1660
#1661
#1662
#1663
#1664
#1669
#1670
#1671
#1672
#1672
#1673
#1674
#1675
#1676
#1677
#1678
#168
#1680
#1682
#1682
#1683
#1683
#1684
#1685
#1686
#1687
#1688
#1689
#1690
#1691
#1692
#1693
#1694
#1695
#1695
#1696
#1697
#1698
#1700
#1701
#1702
#1703
#1703
#1704
#1705
#1706
#1707
#1709
#1709
#1710
#1711
#1712
#1713
#1714
#1715
#1715
#1716
#1717
#1718
#1719
#1719
#1721
#1721
#1722
#1722
#1724
#1725
#1725
#1726
#1727
#1729
#1730
#1731
#1733
#1733
#1734
#1735
#1736
#1737
#1738
#1739
#1740
#1741
#1742
#1743
#1744
#1745
#1746
#1747
#1748
#1749
#1750
#1751
#1753
#1754
#1755
#1756
#1757
#1758
#1759
#1760
#1761
#1763
#1764
#1765
#1766
#1767
#1767
#1769
#1769
#1770
#1771
#1773
#1774
#1774
#1775
#1775
#1776
#1781
#1782
#1783
#1784
#1785
#1788
#1789
#1790
#1791
#1792
#1793
#1794
#1795
#1795
#1796
#1797
#1798
#1799
#1800
#1801
#1802
#1802
#1803
#1804
#1805
#1806
#1807
#1807
#1808
#1809
#1809
#181
#1810
#1810
#1811
#1815
#1816
#1816
#1817
#1818
#1823
#1824
#1825
#1825
#1826
#1827
#1827
#1828
#1829
#1830
#1831
#1832
#1833
#1834
#1835
#1836
#1837
#1838
#1839
#1840
#1841
#1841
#1842
#1843
#1844
#1845
#1846
#1847
#1847
#1849
#1853
#1854
#1855
#1855
#1856
#1857
#1858
#1860
#1860
#1861
#1862
#1863
#1864
#1867
#1868
#1869
#1870
#1872
#1873
#1874
#1877
#1878
#1878
#1879
#188
#1880
#1881
#1881
#1882
#1883
#1883
#1884
#1885
#1885
#1886
#1886
#1887
#1887
#1888
#1889
#1891
#1892
#1893
#1894
#1897
#1898
#1899
#1899
#1900
#1901
#1903
#1904
#1905
#1906
#1907
#1908
#1908
#1909
#1909
#1910
#1910
#1911
#1911
#1912
#1912
#1913
#1914
#1915
#1916
#1917
#1917
#1918
#1918
#1919
#1919
#1920
#1920
#1921
#1922
#1923
#1924
#1925
#1927
#1927
#1928
#1929
#1930
#1931
#1932
#1934
#1935
#1936
#1936
#1937
#1938
#1939
#1941
#1942
#1943
#1944
#1944
#1945
#1946
#1947
#1948
#1949
#1950
#1950
#1951
#1952
#1953
#1953
#1954
#1954
#1955
#1956
#1957
#1957
#1958
#1959
#1959
#196
#1960
#1962
#1963
#1964
#1965
#1966
#1967
#1968
#197
#1970
#1971
#1972
#1972
#1973
#1973
#1975
#1976
#1976
#1977
#1977
#1978
#1978
#1979
#198
#1980
#1981
#1982
#1983
#1984
#1985
#1986
#1986
#1987
#1988
#1988
#1989
#199
#1990
#1990
#1991
#1991
#1992
#1993
#1993
#1994
#1995
#1996
#1997
#1998
#1999
#1999
#2000
#2003
#2004
#2004
#2005
#2006
#2007
#2008
#2009
#2009
#2010
#2012
#2013
#2014
#2017
#2017
#2018
#2019
#202
#2020
#2021
#2022
#2023
#2024
#2025
#2026
#2027
#2028
#2029
#2030
#2031
#2032
#2033
#2034
#2035
#2036
#2037
#2037
#2038
#2039
#2040
#2040
#2041
#2042
#2043
#2044
#2045
#2045
#2046
#2047
#2048
#2049
#2049
#205
#2050
#2051
#2052
#2053
#2054
#2055
#2056
#2057
#2058
#2058
#2059
#2059
#206
#2060
#2061
#2062
#2063
#2064
#2065
#2066
#2067
#2068
#2069
#2069
#2070
#2071
#2072
#2073
#2074
#2075
#2076
#2077
#2077
#2078
#2078
#2079
#208
#2080
#2081
#2082
#2083
#2084
#2085
#2085
#2086
#2087
#2088
#2089
#2089
#209
#2090
#2092
#2093
#2094
#2094
#2096
#2097
#2098
#2099
#210
#2100
#2101
#2102
#2103
#2104
#2105
#2106
#2107
#2108
#2109
#2110
#2111
#2112
#2113
#2114
#2115
#2116
#2117
#2118
#2119
#2119
#2120
#2121
#2122
#2124
#2125
#2126
#2128
#2129
#2129
#2130
#2132
#2133
#2134
#2135
#2136
#2138
#2138
#2139
#214
#2140
#2141
#2142
#2143
#2144
#2145
#2146
#2147
#2148
#2149
#2150
#2151
#2152
#2153
#2154
#2155
#2156
#2157
#2159
#2161
#2161
#2162
#2163
#2164
#2164
#2165
#2166
#2166
#2167
#2167
#2169
#2170
#2171
#2172
#2173
#2173
#2174
#2175
#2176
#2177
#2178
#2179
#2180
#2181
#2182
#2183
#2184
#2185
#2185
#2186
#2186
#2187
#2187
#2188
#219
#2190
#2191
#2192
#2193
#2194
#2195
#2195
#2196
#2197
#2197
#2198
#2198
#2199
#2200
#2201
#2202
#2203
#2205
#2205
#2206
#2207
#2208
#2210
#2211
#2211
#2212
#2213
#2213
#2214
#2215
#2216
#2217
#2218
#2218
#2219
#2221
#2221
#2222
#2223
#2223
#2224
#2225
#2226
#2227
#2228
#223
#2231
#2232
#2234
#2235
#2235
#2236
#2236
#2237
#2237
#2238
#224
#2240
#2240
#2241
#2242
#2243
#2244
#2244
#2245
#2246
#2246
#2247
#2247
#2248
#2248
#2249
#2250
#2250
#2251
#2251
#2252
#2253
#2254
#2255
#2256
#2257
#2258
#2259
#2260
#2261
#2262
#2263
#2264
#2265
#2266
#2266
#2267
#2268
#2269
#227
#2270
#2271
#2272
#2273
#2274
#2275
#2276
#2277
#2278
#2279
#2279
#2280
#2282
#2283
#2283
#2285
#2286
#2287
#2288
#2289
#2290
#2291
#2292
#2292
#2293
#2294
#2295
#2295
#2296
#2298
#2299
#2300
#2301
#2302
#2303
#2304
#2305
#2306
#2307
#2308
#2309
#2310
#2312
#2313
#2313
#2314
#2315
#2315
#2316
#2316
#2317
#2318
#2319
#2320
#2321
#2322
#2323
#2323
#2324
#2325
#2327
#2328
#2329
#2329
#2331
#2331
#2332
#2333
#2334
#2335
#2336
#2338
#2339
#2340
#2340
#2341
#2342
#2343
#2343
#2344
#2345
#2345
#2346
#2347
#2348
#2349
#2351
#2352
#2353
#2354
#2355
#2356
#2357
#2358
#2359
#2360
#2360
#2361
#2361
#2363
#2364
#2365
#2366
#2367
#2367
#2368
#2368
#2369
#2369
#2370
#2370
#2371
#2372
#2372
#2373
#2374
#2375
#2376
#2377
#2378
#2379
#2380
#2381
#2382
#2383
#2384
#2385
#2386
#2387
#2388
#2389
#2390
#2391
#2392
#2393
#2394
#2395
#2396
#2397
#2398
#2399
#2400
#2401
#2401
#2402
#2403
#2404
#2404
#2405
#2406
#2406
#2407
#2408
#2409
#2409
#241
#2410
#2411
#2412
#2413
#2414
#2415
#2415
#2416
#2416
#2417
#2418
#2418
#2419
#2419
#242
#2420
#2420
#2421
#2421
#2422
#2422
#2423
#2425
#2426
#2427
#2428
#2429
#2429
#2430
#2430
#2431
#2431
#2432
#2433
#2434
#2435
#2436
#2437
#2438
#2439
#244
#2440
#2440
#2441
#2441
#2443
#2444
#2445
#2446
#2447
#2448
#2449
#2450
#2451
#2452
#2453
#2453
#2454
#2454
#2455
#2455
#2457
#2458
#2459
#2459
#246
#2460
#2461
#2461
#2462
#2463
#2464
#2465
#2465
#2466
#2467
#2467
#2468
#2469
#2470
#2471
#2472
#2472
#2473
#2473
#2474
#2474
#2475
#2475
#2476
#2476
#2477
#2478
#2479
#248
#2480
#2483
#2484
#2484
#2485
#2485
#249
#2490
#2491
#2492
#2493
#2493
#2494
#2495
#2495
#2496
#2496
#2497
#2497
#2498
#2498
#2499
#2500
#2501
#2502
#2503
#2503
#2504
#2505
#2506
#2507
#2507
#2508
#2508
#2509
#251
#2511
#2512
#2513
#2514
#2515
#2516
#2516
#2517
#2517
#2518
#2519
#2519
#252
#2522
#2522
#2523
#2524
#2525
#2527
#2527
#2528
#2529
#253
#2530
#2531
#2532
#2533
#2534
#2534
#2535
#2535
#2536
#2536
#2537
#2537
#2539
#2539
#254
#2540
#2540
#2541
#2541
#2544
#2545
#2545
#2547
#2548
#2548
#2549
#2550
#2550
#2552
#2552
#2553
#2553
#2554
#2554
#2555
#2555
#2556
#2556
#2557
#2558
#2558
#2559
#2559
#2560
#2560
#2562
#2562
#2564
#2564
#2565
#2566
#2566
#2567
#2567
#2568
#2568
#2569
#2569
#257
#2571
#2572
#2572
#2573
#2574
#2574
#2575
#2575
#2576
#2576
#2577
#2577
#2578
#2578
#2579
#2579
#258
#2580
#2580
#2581
#2582
#2582
#2583
#2583
#2584
#2585
#2585
#2586
#2587
#2587
#2588
#2589
#2589
#2590
#2591
#2592
#2594
#2594
#2595
#2596
#2596
#2597
#2597
#2598
#2598
#2599
#2599
#260
#2600
#2600
#2601
#2601
#2602
#2602
#2603
#2603
#2604
#2604
#2605
#2605
#2606
#2606
#2607
#2607
#2608
#2608
#2609
#2609
#2610
#2610
#2611
#2611
#2612
#2612
#2613
#2613
#2614
#2615
#2616
#2617
#2617
#2618
#2618
#2619
#2620
#2620
#263
#266
#267
#269
#272
#274
#275
#276
#277
#278
#282
#283
#284
#285
#286
#287
#289
#29
#290
#291
#292
#293
#294
#295
#296
#297
#299
#300
#301
#302
#303
#304
#305
#306
#308
#309
#310
#311
#312
#313
#314
#315
#316
#317
#317
#318
#319
#320
#321
#322
#323
#324
#324
#325
#327
#330
#331
#332
#333
#334
#336
#337
#337
#338
#339
#341
#342
#343
#343
#344
#345
#346
#347
#348
#349
#350
#351
#352
#352
#353
#354
#355
#356
#357
#358
#359
#361
#363
#364
#365
#366
#367
#367
#368
#369
#372
#373
#374
#375
#376
#377
#378
#379
#381
#382
#383
#384
#386
#388
#389
#390
#391
#392
#393
#394
#395
#396
#397
#398
#399
#401
#402
#402
#403
#404
#405
#406
#407
#408
#409
#410
#411
#412
#413
#414
#415
#416
#417
#418
#419
#420
#421
#422
#423
#424
#425
#426
#427
#427
#429
#430
#431
#432
#433
#434
#435
#436
#437
#440
#441
#442
#443
#444
#445
#446
#447
#448
#449
#450
#451
#452
#452
#453
#455
#456
#457
#459
#460
#461
#462
#463
#464
#465
#466
#467
#469
#470
#473
#474
#475
#476
#477
#478
#479
#480
#481
#482
#483
#484
#486
#486
#487
#489
#490
#492
#493
#494
#495
#496
#497
#498
#499
#502
#504
#505
#507
#508
#509
#510
#511
#513
#514
#515
#516
#517
#519
#520
#521
#521
#522
#523
#524
#525
#526
#527
#529
#530
#531
#532
#533
#534
#535
#536
#538
#539
#540
#541
#542
#544
#545
#546
#547
#548
#550
#551
#552
#553
#554
#555
#556
#557
#558
#561
#562
#563
#564
#564
#566
#567
#571
#573
#574
#575
#576
#577
#578
#579
#580
#582
#583
#584
#585
#586
#587
#588
#589
#590
#590
#591
#592
#594
#596
#598
#599
#605
#606
#607
#608
#609
#610
#611
#612
#613
#615
#615
#616
#617
#618
#620
#622
#623
#624
#624
#625
#626
#627
#629
#630
#631
#632
#633
#634
#635
#636
#637
#638
#639
#640
#641
#642
#643
#644
#645
#646
#648
#649
#649
#650
#651
#652
#653
#653
#654
#655
#655
#656
#657
#659
#660
#661
#663
#664
#665
#666
#667
#668
#669
#670
#671
#672
#673
#674
#675
#676
#677
#678
#679
#680
#681
#682
#683
#684
#685
#686
#687
#689
#69
#690
#691
#692
#693
#694
#695
#697
#698
#699
#700
#701
#702
#703
#704
#705
#706
#707
#708
#709
#710
#711
#714
#715
#715
#717
#718
#719
#720
#721
#722
#723
#723
#726
#727
#728
#730
#730
#731
#734
#735
#736
#737
#738
#740
#742
#743
#743
#744
#745
#746
#747
#748
#749
#752
#752
#753
#754
#755
#756
#757
#758
#759
#760
#762
#764
#765
#767
#768
#769
#77
#772
#773
#774
#776
#777
#778
#779
#78
#780
#781
#782
#782
#783
#784
#785
#786
#787
#788
#789
#79
#790
#792
#793
#794
#798
#800
#801
#802
#803
#804
#805
#807
#807
#808
#809
#810
#811
#812
#813
#814
#815
#816
#816
#817
#818
#819
#820
#822
#823
#824
#825
#826
#826
#827
#829
#830
#831
#832
#832
#833
#834
#835
#835
#836
#836
#837
#838
#839
#840
#840
#841
#842
#843
#844
#845
#846
#849
#851
#852
#853
#854
#855
#856
#858
#859
#860
#861
#862
#863
#864
#864
#865
#866
#867
#870
#870
#872
#872
#873
#875
#876
#877
#878
#879
#880
#880
#881
#882
#883
#884
#885
#886
#887
#888
#889
#890
#891
#892
#894
#895
#896
#896
#897
#898
#899
#900
#901
#902
#903
#904
#905
#905
#906
#907
#908
#908
#910
#911
#912
#913
#914
#915
#916
#917
#918
#919
#919
#920
#921
#921
#922
#923
#924
#925
#926
#927
#928
#929
#930
#930
#931
#931
#933
#934
#935
#936
#937
#938
#939
#940
#941
#941
#942
#943
#944
#945
#946
#947
#948
#948
#950
#951
#952
#953
#954
#955
#956
#958
#961
#962
#963
#963
#964
#965
#966
#967
#968
#969
#970
#971
#972
#973
#974
#975
#976
#976
#977
#978
#979
#979
#980
#982
#983
#984
#985
#986
#986
#987
#988
#989
#990
#991
#992
#993
#994
#995
#996
#997
#998
#999
0.23.0
0.23.1
0.23.2
0.25.0
0.27.0
0.27.1
0.29.0
0.31.0
0.33.0
0.33.1
0.35.0
0.35.1
0.37.0
0.39.0
0.40.0
0.41.0
0.41.0rc1
0.41.0rc2
0.41.0rc3
0.42.0
0.42.0dev
0.42.0rc0
0.42.0rc1
0.42.0rc2
0.43.0
0.43.0rc0
0.43.0rc1
0.43.0rc2
0.43.0rc3
0.43.0rc4
0.44.0
0.44.0dev
0.44.0rc0
0.44.0rc1
0.44.0rc2
0.44.0rc3
0.44.0rc4
0.44.0rc5
0.45.0
0.45.0dev
0.45.0rc0
0.45.0rc1
0.45.0rc2
0.45.0rc3
0.46.0
0.46.0dev
0.46.0rc0
0.46.0rc1
0.46.0rc2
0.46.1
0.47.0
0.47.0dev
0.47.0rc0
0.47.0rc1
0.47.0rc2
0.48.0dev
-
ad8cd63847
Share one CUDA encoder per IQ family (#2615)
main
Chenjie LuoandClaude Opus 5.5
2026-10-01 12:14:42 -07:00 -
99338d57fe
Deploying to gh-pages from @ NVIDIA/Model-Optimizer@fadbf74d31 🚀
gh-pages
danielkorzekwa
2026-10-01 19:06:06 +00:00 -
5a8c939898
Fix Megatron vLLM fakequant export state
kinjal/vllm_wrapper
Kinjal Patel
2026-10-01 19:05:51 +00:00 -
fadbf74d31
Make the QAT/QAD guide the central place for concepts, background, and framework selection (#2590)
Daniel Korzekwa
2026-10-01 20:56:57 +02:00 -
8009e713c0
docs(qwen3.6): explain why the penalty speeds up SciCode and slows GPQA
pull-request/2606
kmorabia/scicode-verbosity-mitigation
Keval MorabiaandClaude Opus 5
2026-10-01 11:43:34 -07:00 -
4f62d418f4
Use TensorRT optimization level 0 in Torch ONNX example tests (#2619)
Ajinkya Rasane
2026-10-01 14:32:18 -04:00 -
f2b9ecbd8a
llm_sparsity: store the local copy as local_checkpoint_path, the ID as hub_model_id
shengliangx/model-download-5-llm-sparsity
Shengliang Xu
2026-10-01 01:00:44 +00:00 -
9c28fab4b9
llm_sparsity: export from a local checkpoint
Shengliang Xu
2026-09-30 21:36:27 +00:00 -
d765177226
megatron_bridge: store the local copy as local_checkpoint_path, the ID as hub_model_id
shengliangx/model-download-4-megatron
Shengliang Xu
2026-10-01 01:00:13 +00:00 -
e9820f564f
Megatron export: copy non-model files from local disk only
Shengliang Xu
2026-09-30 21:36:06 +00:00 -
9ce3570eb9
hf_ptq: store the local copy as local_checkpoint_path, the ID as hub_model_id
shengliangx/model-download-3-hf-ptq
Shengliang Xu
2026-10-01 00:59:30 +00:00 -
cf7c4438fe
hf_ptq: load from a local checkpoint and carry its files verbatim
Shengliang Xu
2026-09-30 21:30:35 +00:00 -
137a128a8c
HF exporters write off-index safetensors and non-model files
shengliangx/model-download-2-hf-exporters
Shengliang Xu
2026-09-30 21:25:11 +00:00 -
2d71a30ff7
Name ensure_local_checkpoint's results hub_model_id and local_checkpoint_path
shengliangx/model-download-1-hf-checkpoint-utils
Shengliang Xu
2026-10-01 00:58:17 +00:00 -
79570d639d
Add ensure_local_checkpoint and copy_non_model_files
Shengliang Xu
2026-09-30 21:20:23 +00:00 -
439ad4bcf5
Add vLLM NVFP4 MLA KV-cache fake quant; fix FP8 and CUDA-graph serving
pull-request/2620
Shiyang Chen
2026-09-30 17:51:37 -07:00 -
a7efcc9f55
Merge branch 'main' into chenjiel/iq-kernel-dedup
pull-request/2615
chenjiel/iq-kernel-dedup
Chenjie Luo
2026-10-01 11:01:47 -07:00 -
4fa387bd0a
test(export): consolidate reverse conversion regression coverage
weimingc/fix-export-names
pull-request/2508
weimingc
2026-10-01 10:49:40 -07:00 -
9c1cf80f1b
test(megatron_bridge): cover context parallelism in the VLM QAD test (#2592)
Keval MorabiaandClaude Opus 5
2026-10-01 23:05:55 +05:30 -
2ad0e70ae8
Address review: free teacher fp32 copy early, input-only skip_lm_loss, dtype-tied ghost floor
pull-request/2459
aanoosheh/topk-kd-topp-ghost
Asha AnooshehandClaude Opus 5.5
2026-10-01 19:12:44 +02:00 -
2beb70ffab
Merge main and resolve the GGML format registry conflict
pull-request/2516
hungyuehc/q8-0-quantization
Hung-Yueh Chiang
2026-10-01 09:06:54 -07:00 -
9e1ede83f8
refactor(speculative): project the base distribution only where a loss reads it, family-wide
haoguo/dflash-training-perf
h-guo18andClaude Opus 5.5
2026-10-01 15:16:11 +00:00 -
fe130f8bfe
docs(speculative): trim this PR's comments down to the why
h-guo18andClaude Opus 5.5
2026-10-01 13:59:21 +00:00 -
c1857f05d5
Hard error on deprecated KDConfig args for simplicity
Asha Anoosheh
2026-10-01 15:58:11 +02:00 -
04fdfbd61e
revert(speculative): keep ddp_find_unused_parameters true in the Gemma-4 DSpark recipe
h-guo18andClaude Opus 5.5
2026-10-01 13:46:56 +00:00 -
8f85d402da
Merge branch 'main' into lilicorr-streaming-cal
pull-request/2589
mrusanovsky
2026-10-01 16:24:02 +03:00 -
923044fc7d
review: cast the captured hidden state before the final norm
mrusanovsky
2026-10-01 06:15:52 -07:00 -
f2449a2747
fix(speculative): keep the DFlash draft's RoPE frequencies in fp32 on a fresh convert
haoguo/dflash-rope-fp32
h-guo18andClaude Opus 5.5
2026-10-01 12:47:06 +00:00 -
7d9e07d14b
Count only added lines toward the PR size budget in AGENTS.md (#2616)
Chenjie LuoandClaude Opus 5.5
2026-10-01 00:03:48 -07:00 -
5bf43143c9
Decode the IQ formats on CUDA
pull-request/2604
chenjiel/iq-export-reuse-cuda-decode
Chenjie LuoandClaude Opus 5.5
2026-10-01 06:24:56 +00:00 -
7dfc5aee1f
Export IQ weights from the payload fake quant already packed
Chenjie LuoandClaude Opus 5.5
2026-10-01 06:22:01 +00:00 -
3947a86045
Report packer input errors in the original order
Chenjie LuoandClaude Opus 5.5
2026-10-01 06:36:28 +00:00 -
6956a28b66
Share one CUDA encoder per IQ family
Chenjie LuoandClaude Opus 5.5
2026-10-01 06:21:48 +00:00 -
389eac6867
fix(export): preserve scoped experts and clarify conversion fallback
weimingc
2026-09-30 22:57:05 -07:00 -
aa89722d38
[6/6] Add the IQ1_M CUDA encoder and register the format (#2595)
Chenjie LuoandClaude Opus 5.5
2026-09-30 22:23:48 -07:00 -
333ace1bc9
Fix multi-turn synthetic generation and mark incomplete outputs (#2571)
sychen52
2026-09-30 17:37:08 -07:00 -
8d489b1ae7
docs(qwen3.6): show the 338 x 8 pooling and point at the shipped config line
Keval MorabiaandClaude Opus 5
2026-09-30 17:34:15 -07:00 -
bc5d2c3610
Update roadmap link in README.md (#1700)
Trenton Starkey - Product @ NVIDIA
2026-09-30 17:06:47 -07:00 -
2fdf4069a2
docs(qwen3.6): draw the token means from the same source as the table above
Keval MorabiaandClaude Opus 5
2026-09-30 16:59:10 -07:00 -
5c3433bcc9
docs(qwen3.6): show that the penalty removes the tail, not the verbosity
Keval MorabiaandClaude Opus 5
2026-09-30 16:45:10 -07:00 -
94fe9a7bf3
docs(qwen3.6): drop a causal over-read and align the tau2 config comment
Keval MorabiaandClaude Opus 5
2026-09-30 16:11:57 -07:00 -
64bb7fdf7e
Keep quantization scales aligned with weight-specific export renames
weimingc
2026-09-30 15:56:06 -07:00 -
3a057325e5
docs(qwen3.6): resolve the accuracy-headroom puzzle and signpost the config
Keval MorabiaandClaude Opus 5
2026-09-30 15:55:35 -07:00 -
472ad85ba0
docs(qwen3.6): correct the BF16 row and hedge the capped definition
Keval MorabiaandClaude Opus 5
2026-09-30 15:41:32 -07:00 -
e7b41b7823
inline _install_kpool_cache_hooks
pull-request/2552
Shiyang Chen
2026-09-30 14:55:17 -07:00 -
968241bba1
fix(export): align tensor names and wildcard exclusions
weimingc
2026-09-30 14:54:21 -07:00 -
4e949851ce
docs: add single-GPU PTQ announcement blog
fridah/blog-single-gpu-ptq
Fridah-nvandClaude Opus 5.5
2026-09-28 19:18:17 +00:00 -
948d3d95f4
docs(qwen3.6): add the decode-time fix for SciCode non-termination
Keval MorabiaandClaude Opus 5
2026-09-30 14:28:41 -07:00 -
5ca061e0f1
fix(export): handle weight-specific merge conversion patterns
weimingc
2026-09-30 14:20:00 -07:00 -
a68581654b
docs(ptq): select PyTorch base image and compatible Transformers
pull-request/2605
docs/ptq-pytorch-base-image
Chad Voegele
2026-09-30 21:06:28 +00:00 -
e53e9911e0
Add fake quantization of the sparse-attention indexer query
Shiyang Chen
2026-09-30 12:09:07 -07:00 -
50d697647e
review: offline logit reconstruction casts to the lm_head dtype; log calibration as a KL
mrusanovsky
2026-09-30 12:07:39 -07:00 -
3c0cfd1486
Merge latest main and IQ1_M registration into Q8_0 codec
Hung-Yueh Chiang
2026-09-30 12:04:00 -07:00 -
470a97944b
Address review: exporter fallback, Nemotron VL detection, changelog
shengliangx/deprecating-trtllm-model_type
pull-request/2569
Shengliang Xu
2026-09-30 18:48:21 +00:00 -
0cf35019e1
Merge remote-tracking branch 'origin-oss/main' into shengliangx/deprecating-trtllm-model_type
Shengliang Xu
2026-09-30 18:44:20 +00:00 -
6220daf4b2
Limit indexer K-cache fake quantization to DeepSeek-V4 and GLM-5.3-Flash
Shiyang Chen
2026-09-29 15:16:24 -07:00 -
4754d4a8dc
vllm_serve: move quantizers to the GPU before CUDA graph capture
Shiyang Chen
2026-09-26 17:13:57 -07:00 -
7539d89669
Address review: skip disabled GLM-5.3-Flash kpool hooks, explain DSA-only freeze
Shiyang Chen
2026-09-26 16:15:27 -07:00 -
a7a27fb1d9
Add sparse-attention indexer K-cache fake quantization
Shiyang Chen
2026-09-26 11:24:46 -07:00 -
49f8d1eda8
Add the IQ1_M CUDA encoder and register the format
Chenjie LuoandClaude Opus 5.5
2026-09-29 22:46:09 +00:00 -
e5b63320ab
[5/6] Add the IQ1_M codec (#2513)
Chenjie LuoandClaude Opus 5.5
2026-09-30 09:45:48 -07:00 -
c35f0aa92a
launcher: end-to-end DFlash reference run for Qwen3-8B
yeyu/specdec-e2e-reference-run
Ye YuandClaude Opus 5
2026-09-30 09:16:44 -07:00 -
b8cbcab8a3
skills(specdec): add DFlash2 and LiLiCorr algorithm sheets
yeyu/specdec-skill-sheets-dflash2-lilicorr
pull-request/2602
Ye YuandClaude Opus 5
2026-09-30 09:10:45 -07:00 -
57fc7032bd
Merge branch 'main' into aanoosheh/topk-kd-topp-ghost
Asha AnooshehandClaude Opus 5.5
2026-09-30 12:10:33 +02:00 -
51a8b606f1
Clarify that the ghost token is always appended in CHANGELOG
Asha AnooshehandClaude Opus 5.5
2026-09-30 12:10:05 +02:00 -
cb51cac889
Checkpoint ONNX sensitivity scores
pull-request/2240
gcunhase
2026-09-30 01:04:48 +00:00 -
b4cabf1a22
[OMNIML-5935] Fix hf_ptq overwriting valid pad_token with eos_token
pull-request/2597
fix/hf-ptq-pad-token-overwrite
Zhiyu ChengandClaude Sonnet 4.6
2026-09-29 17:43:00 -07:00 -
98e108a681
Keep the int4_awq and w4a8_awq_beta hf_ptq cases
pull-request/2596
chenjiel/trim-hf-ptq-example-tests
Chenjie LuoandClaude Opus 5.5
2026-09-29 23:10:36 +00:00 -
a700005703
Keep fp8 kv none and use nvfp4 KV for the autoquant case
Chenjie LuoandClaude Opus 5.5
2026-09-29 23:06:54 +00:00 -
c6e60a12a7
Trim redundant hf_ptq example test cases
Chenjie LuoandClaude Opus 5.5
2026-09-29 22:58:05 +00:00 -
4826fb7554
Use minimum TensorRT build optimization in diffusers tests
pull-request/2594
ajrasane/reduce-diffusers-test-compute
ajrasane
2026-09-29 22:47:01 +00:00 -
2cf6f1f06a
cleanup
Kinjal Patel
2026-09-29 22:44:05 +00:00 -
def89fe169
Clean up vLLM fakequant launcher compatibility code
Kinjal Patel
2026-09-29 22:30:53 +00:00 -
d8007ff792
Fix fakequant launcher model detection
Kinjal Patel
2026-09-17 22:39:57 +00:00 -
426f5a9828
updated for 0.29.0
Kinjal Patel
2026-09-17 22:16:11 +00:00 -
4fa880387d
cleanup
Kinjal Patel
2026-09-17 20:33:45 +00:00 -
6f67fa993f
README update
Kinjal Patel
2026-09-17 20:07:58 +00:00 -
29b44fc4ce
removed combined mlflow export
Kinjal Patel
2026-09-17 19:25:15 +00:00 -
b779166ace
cleanup
Kinjal Patel
2026-09-17 19:18:31 +00:00 -
0edad0e830
bug fix
Kinjal Patel
2026-09-16 03:26:46 +00:00 -
00f8a54eb7
Cleanup
Kinjal Patel
2026-09-15 19:16:42 +00:00 -
5626588d33
mlflow combined
Kinjal Patel
2026-09-09 21:16:34 +00:00 -
c6e507e496
added triton as default backend
Kinjal Patel
2026-09-04 21:13:49 +00:00 -
7d54a9267c
install as app
Kinjal Patel
2026-08-28 22:46:51 +00:00 -
5be1c4a788
testing vllm interface
Kinjal Patel
2026-08-28 22:19:05 +00:00 -
1e5f90e3dd
allow vllm fq to run on unquantized layers in quantized models
Kinjal Patel
2026-08-19 22:44:37 +00:00 -
5a2dbdb68c
fixed for GLM
Kinjal Patel
2026-09-24 01:56:19 +00:00 -
f644fab448
Resolve vLLM compilation marker owner
Kinjal Patel
2026-09-23 02:01:23 +00:00 -
f0a85d559c
Reduce diffusers example-test compute with tiny models
ajrasane
2026-09-29 21:41:44 +00:00 -
97d0058bcb
fix(export): guard quantized merges and align module names
weimingc
2026-09-29 14:32:56 -07:00 -
ad9ea97a4b
Fix vLLM compilation guard for models without marker (#2518)
kinjalpatel27
2026-09-29 14:28:32 -07:00 -
3091b8ff69
[4/5] Add the IQ2_S CUDA encoder and register the format (#2565)
Chenjie LuoandClaude Opus 5.5
2026-09-29 13:35:38 -07:00 -
1576f7d4ad
Increase diffusers example-test timeout to 60 minutes (#2591)
Ajinkya Rasane
2026-09-29 15:15:05 -04:00 -
078798c323
sensitivity: redirect stdout during probe to silence ORT calibrator print()s
gcunhaseandmodelopt-fix-agent-bot
2026-09-29 18:56:57 +00:00 -
834c90d7a1
Fix NVFP4 fake quant zeroing blocks with small scales (#2549)
sychen52andClaude Opus 5.5
2026-09-29 10:31:31 -07:00 -
99179109c0
Addressed bot review comments
rohjoshi/scale-learning-blog
pull-request/2453
Rohan Joshi
2026-09-29 17:13:23 +00:00 -
498436cb44
First commit
Rohan Joshi
2026-09-16 23:15:16 +00:00 -
461921aa18
Accept sensitivity log_level in score
gcunhase
2026-09-29 16:53:58 +00:00 -
3fa3709d09
sensitivity: add --log_level, quiet per-probe sub-log noise
gcunhaseandmodelopt-fix-agent-bot
2026-09-29 16:45:00 +00:00 -
dc1d3363ed
review: state the NoPE fix by outcome; cover every candidate in the calibration loss test
mrusanovsky
2026-09-29 08:52:15 -07:00