mirror of
https://github.com/FlashML-org/FreeToken.git
synced 2026-11-22 02:36:36 +08:00
* feat(mm): serve image input on the Qwen VL families - frontend: OpenAI image_url, Anthropic image blocks and Responses input_image (http(s), base64, gated file://) become MMItems in the tokenizer workers; image spans carry content pseudo ids so the radix cache keys them by content; mrope positions are precomputed per request - registry: EncoderSpec per family (Qwen3-VL, Qwen3.6, Qwen3.8-Flash-Next register a vision encoder serving image) and ModelSpec.mm_processor; EngineConfig.active_encoders decides what a process builds, served_modalities what it accepts; --text-model-only, --mm-disable, --mm-encoder-weights, --mm-embed-cache-device, --mm-max-pixels - engine and scheduler: per-request encoder-cache row ownership, chunk planning on the batch (jobs, gather rows, scatter rows), items encoded right before the LM forward and gathered into batch.mm_embeds - models: the Qwen VL vision tower on the text tower's TP layers with DeepStack, block weights streamed from pinned host banks by default, merger outputs parked on the host between blocks; SupportsMultimodal (encode, place_encoder_weights) and a shared embed_input_ids - server: image inputs rejected with a client-facing reason when no vision encoder is served; GET /v1/stats reports model.input_modalities * feat(mm): replace the pixel cap with a token budget and processor kwargs - --image-min-tokens / --image-max-tokens replace --mm-max-pixels; MMProcessor.get_mm_processor_kwargs turns the MultimodalConfig into the processor call's arguments, the default passes mm.processor_kwargs through and Qwen VL converts the budget to pixel areas in size.shortest_edge / longest_edge - --mm-processor-kwargs hands a JSON object to the checkpoint's image processor call, applied after the token budget - MMProcessor is an ABC that owns the model path, the config and the lazily loaded image processor; get_mm_processor returns None when --mm-disable leaves no encoder to serve - the frontend tokenizer, the tokenizer worker and the offline LLM build one processor each and hand it to TokenizeManager, so the tokenize message carries no per-request pixel cap
37 lines
764 B
Python
37 lines
764 B
Python
from __future__ import annotations
|
|
|
|
from dataclasses import dataclass
|
|
from typing import TYPE_CHECKING, List
|
|
|
|
import torch
|
|
|
|
if TYPE_CHECKING:
|
|
from freetoken.core import SamplingParams
|
|
|
|
from .prefill import ChunkedReq
|
|
|
|
|
|
@dataclass
|
|
class PendingReq:
|
|
uid: int
|
|
input_ids: torch.Tensor
|
|
sampling_params: SamplingParams
|
|
chunked_req: ChunkedReq | None = None
|
|
mm_items: list | None = None
|
|
mrope_positions_full: torch.Tensor | None = None
|
|
mrope_delta: int = 0
|
|
|
|
@property
|
|
def input_len(self) -> int:
|
|
return len(self.input_ids)
|
|
|
|
@property
|
|
def output_len(self) -> int:
|
|
return self.sampling_params.max_tokens
|
|
|
|
|
|
@dataclass
|
|
class ScheduleResult:
|
|
reqs: List[PendingReq]
|
|
output_indices: List[torch.Tensor]
|