Production runs one static model behind the OpenAI-compatible API. Every
adapter path and non-OpenAI HTTP surface is unreachable there.
- Delete the LoRA stack: vllm/lora, LoRARequest on the engine wire, the
scheduler, worker, cudagraph and MoE plumbing, LoRA metrics, the lora
CLI group, and the LoRA tests.
- Drop the plugin loader in vllm/plugins. Platform resolution is now
CUDA-only and the general/stat-logger plugin hooks are gone.
- Delete the speech-to-text and realtime entrypoints, the scale-out
render routes, and vllm/config/speech_to_text.py. GenerationTask is
generate-only.
- Cut DecodeBenchConnector and the disagg-prefill doc and image. The
Multi, Offloading and Example connectors stay; production uses the
CPU KV offload tier.
- Delete the ROCm-only fused KDA chunk kernel (2346 lines) with its CMake
arch gate, declarations and op registrations. The CUDA build never
defined the guard and no Python caller existed.
- Delete the orphan Mistral and Pixtral chat fixtures and examples/.
- Fix _check_model: removing the LoRA early return made every request
fail with "The model ... does not exist. Valid aliases: ...".
Verified on an RTX 3090: import sweep reports 1242 modules with 7 known
optional-dependency failures; a live server answers chat, completions,
responses, tokenize and detokenize with zero log errors and 404 for the
removed surfaces; collection yields 8558 tests with the 3 pre-existing
errors; ruff adds no new finding; the incremental _C build succeeds.
Co-authored-by: Hinata <hinata@idihore.id>
Signed-off-by: MOVZX <movzx@yahoo.com>
- Delete csrc/rocm, hipify, the _rocm_C extension and the setup.py rocm
target path; drop the rocm requirement files
- Delete the aiter, rdna3, skinny-gemm and ROCm fused-MoE kernel modules
and retarget the linear kernel registries away from them
- Reduce platforms/rocm.py to the GCN architecture predicates that live
code still imports, and delete the RocmPlatform class and plugin
- Drop the VLLM_SERVER_DEV_MODE debug surface: reset, collective_rpc,
server_info and sleep routes plus the log_error_stack flag
- Delete lm-format-enforcer, lark and otel-semantic-conventions-ai pins;
the live server process maps none of them
- Delete tests that only build architectures this fork no longer
registers, and test directories left without any test file
- Restore the tests/v1/kv_connector mock helpers that scheduler tests
import at module level
Co-authored-by: Hinata <hinata@idihore.id>
Signed-off-by: MOVZX <movzx@yahoo.com>
- Restore example connector, test helpers, and dropped patch tests
- Guard MLA metadata import in the example connector
- Point the PLE test at common.ops.ple after upstream #60021 move
- Cut tests for removed features: tools, DRY, epd proxy, STT batch,
diffusion scheduler, INC/Quark PLE cases, mamba-2 mixer
Co-authored-by: Hinata <hinata@idihore.id>
Signed-off-by: MOVZX <movzx@yahoo.com>
- Keep only Qwen3.5, Qwen3.5-MTP and Qwen4Exp model paths; trim the model registry to 41 entries
- Delete idle attention backends (MLA, FlashInfer, ROCm-only, quant backends) and dead GDN CuTeDSL kernels
- Remove inc/quark quant loaders; keep compressed-tensors and Marlin
- Drop the Rust frontend, pooling entrypoints, and idle test suites
- Rebuild kernels for sm_80;sm_86 and strip the extensions
- Verified: full import sweep clean (optional deps only) and a dummy-load Qwen3.5 engine run on RTX 3090 via the Triton GDN path
Co-authored-by: Hinata <hinata@idihore.id>
Signed-off-by: MOVZX <movzx@yahoo.com>
Intake of open upstream PR vllm-project/vllm#58863. Active only with
VLLM_USE_V2_MODEL_RUNNER=1; inert otherwise.
Speculative decoding on Gated DeltaNet snapshotted the full recurrent
state after every draft position and reserved one Mamba block per draft
position. On a 24 GB card that is reserved VRAM we never use, plus
per-step copy traffic.
RecoverSSM, the protocol Kimi-K3 KDA already uses, replaces this:
verify runs from a read-only checkpoint and stores a small per-token
replay record [correction (V), normalized key (K), gate (1)] in fp32;
commit replays only the accepted tokens and writes the recurrent state
once, plus the block-boundary state in align mode. The PLE dilated
short conv compacts its accepted window with the same plan.
Conflict in vllm/models/qwen4_exp/nvidia/mtp.py: our get_top_tokens()
work put LocalArgmaxMixin on Qwen4ExpMTP, the PR adds SupportsReplaySSM
to the same base list. Kept both.
Co-Authored-By: Hinata <hinata@idihore.id>
Signed-off-by: MOVZX <movzx@yahoo.com>
Adds an opt-in backend, --engram-config '{"host_file_gather": true}', that
serves Qwen4Exp PLE rows from the checkpoint files through the page cache
instead of allocating the 47.7 GiB table on GPU or in pinned host memory.
- Qwen4ExpPLEFileGatherEmbedding allocates no table rows; load_weights keeps
the default loader's zero-copy safetensors views of each shard. At bind
(post-load hook) every view must be file-backed per /proc/self/maps (HF
cache symlinks work; heap/pinned/anonymous memory is rejected) and the
shards must cover the whole table.
- Each step, Qwen4ExpModelState.prepare_inputs computes the PLE n-gram ids,
copies them to pinned host memory, syncs the stream once, issues
POSIX_FADV_WILLNEED over the coalesced page ranges of the step's rows,
gathers the rows on the host, and copies them into a fixed device staging
buffer. The forward reads only the staging buffer, so FULL CUDA graph
capture is unaffected.
- Rows stay in the checkpoint dtype (FP8 table stays FP8, dequantized on the
GPU); dummy loads serve zeros; weight reload is rejected; takes precedence
over cpu_offload; rejected on ROCm and with elastic EP.
Co-Authored-By: Hinata <hinata@idihore.id>
Signed-off-by: MOVZX <movzx@yahoo.com>
RecoverSSM's commit-plan and conv-window compaction kernels are model independent.
Move them from models/kimi_k3 into model_executor/layers/mamba/ops/recoverssm.py
(Kimi-K3 keeps its private names as import aliases; kernel source unchanged), add
mamba/recoverssm_utils.py (activation predicate, argument validation) and the
generic spec-row/request mapping helpers to v1/attention/backends/recoverssm_metadata.py.
CacheConfig.use_kda_recoverssm becomes use_recoverssm: the switch now selects RecoverSSM
for any model whose recurrent layers implement it, not only KDA.
No functional change for Kimi-K3.
Signed-off-by: Jürgen Schmied <juergenschmied70@gmail.com>
Co-authored-by: Claude <noreply@anthropic.com>