Please confirm you are human

This browser or connection looks automated. Press and continuously hold the control for 3 seconds to enable Google-hosted web results and, when separately allowed, AI-assisted answers.

A successful check enables 100 search requests. Interactive access does not authorize scraping, systematic collection, or reuse of search output.

Hold with a pointer, or hold Space or Enter.

News

vLLM
docs.vllm.ai > en > latest > api > vllm > models > hy_v4 > nvidia > model

model

3+ hour, 21+ min ago   (68+ words) Inference-only HY V4 model compatible with HuggingFace weights (NVIDIA). One HY V4 decoder layer: MLA attention plus a dense or MoE MLP. Whenconfig.enable_ihc is set the layer runs on hc_mult residual channels and each sub-block is wrapped by an HYV4HCLayer boundary; otherwise it uses…...

vLLM
docs.vllm.ai > en > latest > api > vllm > models > hy_v4 > nvidia > attention

attention

3+ hour, 21+ min ago   (434+ words) MLA attention and lightning indexer for HY V4 (NVIDIA). The per-head learnable sink is supported through.flashmla_sparse, which subclasses the platform's sparse MLA backend to forward attn_sink. The sink is applied by binding the sink-capable backend from.flashmla_sparse; if no backend on this platform…...

vLLM
docs.vllm.ai > en > latest > api > vllm > reasoning > hy_v4_reasoning_parser

hy_v4_reasoning_parser

3+ hour, 21+ min ago   (286+ words) Reasoning extraction for HYV4, on plain data. Count reasoning tokens (0 in no_think). The start token is injected at the END of the prompt, so it is normally absent from token_ids. Only skip a leading start token when it actually appears first (legacy / no…...

vLLM
docs.vllm.ai > en > latest > api > vllm > models > hy_v4 > nvidia > flashmla_sparse

flashmla_sparse

3+ hour, 21+ min ago   (254+ words) Sink-capable FlashMLA sparse backend for HY V4 (NVIDIA). HY V4 adds a per-head learnable attention sink on top of sparse MLA. The vendored FlashMLA kernels already accept an attn_sink argument, but vLLM's shared FLASHMLA_SPARSE backend neither advertises sink support nor forwards the tensor, so…...

vLLM
docs.vllm.ai > en > stable > examples > disaggregated > lmcache

LMCache Examples

3+ mon, 2+ week ago   (81+ words) This folder demonstrates how to use LMCache with vLLM v1 for KV cache offloading, disaggregated prefilling, and KV cache sharing. LMCache integrates with vLLM v1 in two ways: This example demonstrates how to run LMCache with disaggregated prefill using NIXL on a…...

vLLM
docs.vllm.ai > en > latest > api > vllm > distributed > weight_transfer > sharded_rdt_engine

sharded_rdt_engine

6+ day, 22+ hour ago   (1746+ words) Sharded Ray Direct Transport (RDT) weight transfer engine (consumer side). Pulls only the slice each vLLM worker consumes under tensor/expert parallelism, not the full HF-format tensor. One thing at four resolutions, over three lifetimes. FetchKey -- (name, op_chain), "which slice of…...

vLLM
docs.vllm.ai > en > latest > training > weight_transfer > sharded_rdt

Sharded RDT Engine

6+ day, 22+ hour ago   (259+ words) Discovery is expensive, so it happens once, at init_transfer_engine, as a dry run over model.load_weights with every parameter on meta. Nothing is transferred; the engine just records, per leaf module, which slice feeds which destination region. Every later sync is pure…...

vLLM
docs.vllm.ai > en > latest > api > vllm > models > deepseek_v32 > nvidia > glm52_low_latency_gemm

glm52_low_latency_gemm

3+ week, 4+ day ago   (26+ words) vLLM GLM-5.2 decode GEMM selection for unquantized BF16 on SM103. Plan for a weight the walk below cannot reach (a plain nn.Linear )....

vLLM
docs.vllm.ai > en > latest > api > vllm > model_executor > models > glm_ocr

glm_ocr

7+ mon, 3+ day ago   (10+ words) vLLM Inference-only GLM-OCR model compatible with HuggingFace weights....

Google News
docs.vllm.ai > en > latest > api > vllm > models > dots3_note > nvidia > audio

audio

2+ week, 1+ day ago   (22+ words) vLLM _forward_speech_encoder(mel_features, input_seq_lens, audio_sample_lens) ¶ Run the eager speech encoder without server-side slicing/batching....