Skip to content

Commit c99c4c8

Browse files
wangxiyuanYikun
wangxiyuan
and
Yikun
authored
[Doc] Update feature support list (#650)
1. remove Chinese doc. The content is out of data and we don't have enough time to maintain it. 2. Update feature support matrix. Refresh the content and add V1 status. --------- Signed-off-by: wangxiyuan <[email protected]> Signed-off-by: Yikun Jiang <[email protected]> Co-authored-by: Yikun Jiang <[email protected]>
1 parent 3879d9c commit c99c4c8

File tree

3 files changed

+42
-200
lines changed

3 files changed

+42
-200
lines changed

‎docs/source/developer_guide/contributing.zh.md‎

Lines changed: 0 additions & 102 deletions
This file was deleted.

‎docs/source/developer_guide/versioning_policy.zh.md‎

Lines changed: 0 additions & 79 deletions
This file was deleted.

‎docs/source/user_guide/suppoted_features.md‎

Lines changed: 42 additions & 19 deletions
Original file line numberDiff line numberDiff line change
@@ -1,21 +1,44 @@
11
# Feature Support
22

3-
| Feature | Supported | CI Coverage | Guidance Document | Current Status | Next Step |
4-
|--------------------------|-----------|-------------|-------------------|---------------------------|--------------------|
5-
| Chunked Prefill | ❌ | | | NA | Plan in 2025.03.30 |
6-
| Automatic Prefix Caching | ❌ | | | NA | Plan in 2025.03.30 |
7-
| LoRA | ❌ | | | NA | Plan in 2025.06.30 |
8-
| Prompt adapter | ❌ | | | NA | Plan in 2025.06.30 |
9-
| Speculative decoding | ✅ | | | Basic functions available | Need fully test |
10-
| Pooling | ✅ | | | Basic functions available(Bert) | Need fully test and add more models support|
11-
| Enc-dec | ❌ | | | NA | Plan in 2025.06.30|
12-
| Multi Modality | ✅ | | ✅ | Basic functions available(LLaVA/Qwen2-vl/Qwen2-audio/internVL)| Improve performance, and add more models support |
13-
| LogProbs | ✅ | | | Basic functions available | Need fully test |
14-
| Prompt logProbs | ✅ | | | Basic functions available | Need fully test |
15-
| Async output | ✅ | | | Basic functions available | Need fully test |
16-
| Multi step scheduler | ✅ | | | Basic functions available | Need fully test, Find more details at [<u> Blog </u>](https://blog.vllm.ai/2024/09/05/perf-update.html#batch-scheduling-multiple-steps-ahead-pr-7000), [<u> RFC </u>](https://github.com/vllm-project/vllm/issues/6854) and [<u>issue</u>](https://github.com/vllm-project/vllm/pull/7000) |
17-
| Best of | ✅ | | | Basic functions available | Need fully test |
18-
| Beam search | ✅ | | | Basic functions available | Need fully test |
19-
| Guided Decoding | ✅ | | | Basic functions available | Find more details at the [<u>issue</u>](https://github.com/vllm-project/vllm-ascend/issues/177) |
20-
| Tensor Parallel | ✅ | | | Basic functions available | Need fully test |
21-
| Pipeline Parallel | ✅ | | | Basic functions available | Need fully test |
3+
The feature support principle of vLLM Ascend is: **aligned with the vLLM**. We are also actively collaborating with the community to accelerate support.
4+
5+
You can check the [support status of vLLM V1 Engine][v1_user_guide]. Below is the feature support status of vLLM Ascend:
6+
7+
| Feature | vLLM V0 Engine | vLLM V1 Engine | Next Step |
8+
|-------------------------------|----------------|----------------|------------------------------------------------------------------------|
9+
| Chunked Prefill | 🚧 WIP | 🚧 WIP | Functional, waiting for CANN 8.1 nnal package release |
10+
| Automatic Prefix Caching | 🚧 WIP | 🚧 WIP | Functional, waiting for CANN 8.1 nnal package release |
11+
| LoRA | 🟢 Functional | 🚧 WIP | [vllm-ascend#396][multilora], CI needed, working on V1 support |
12+
| Prompt adapter | No plan | 🟡 Planned | Plan in 2025.06.30 |
13+
| Speculative decoding | 🟢 Functional | 🚧 WIP | CI needed; working on V1 support |
14+
| Pooling | 🟢 Functional | 🟢 Functional | CI needed and adapting more models; V1 support rely on vLLM support. |
15+
| Enc-dec | 🔴 NO plan | 🟡 Planned | Plan in 2025.06.30 |
16+
| Multi Modality | 🟢 Functional | 🟢 Functional | [Tutorial][multimodal], optimizing and adapting more models |
17+
| LogProbs | 🟢 Functional | 🟢 Functional | CI needed |
18+
| Prompt logProbs | 🟢 Functional | 🟢 Functional | CI needed |
19+
| Async output | 🟢 Functional | 🟢 Functional | CI needed |
20+
| Multi step scheduler | 🟢 Functional | 🔴 Deprecated | [vllm#8779][v1_rfc], replaced by [vLLM V1 Scheduler][v1_scheduler]) |
21+
| Best of | 🟢 Functional | 🔴 Deprecated | [vllm#13361][best_of], CI needed |
22+
| Beam search | 🟢 Functional | 🟢 Functional | CI needed |
23+
| Guided Decoding | 🟢 Functional | 🟢 Functional | [vllm-ascend#177][guided_decoding] |
24+
| Tensor Parallel | 🟢 Functional | 🟢 Functional | CI needed |
25+
| Pipeline Parallel | 🟢 Functional | 🟢 Functional | CI needed |
26+
| Expert Parallel | 🔴 NO plan | 🟢 Functional | CI needed; No plan on V0 support |
27+
| Data Parallel | 🔴 NO plan | 🟢 Functional | CI needed; No plan on V0 support |
28+
| Prefill Decode Disaggregation | 🟢 Functional | 🟢 Functional | 1P1D available, working on xPyD and V1 support. |
29+
| Quantization | 🟢 Functional | 🟢 Functional | W8A8 available, CI needed; working on more quantization method support |
30+
| Graph Mode | 🔴 NO plan | 🟢 Functional | Functional, waiting for CANN 8.1 nnal package release |
31+
| Sleep Mode | 🟢 Functional | 🟢 Functional | level=1 available, CI needed, working on V1 support |
32+
33+
- 🟢 Functional: Fully operational, with ongoing optimizations.
34+
- 🚧 WIP: Under active development
35+
- 🟡 Planned: Scheduled for future implementation (some may have open PRs/RFCs).
36+
- 🔴 NO plan / Deprecated: No plan for V0 or deprecated by vLLM v1.
37+
38+
[v1_user_guide]: https://docs.vllm.ai/en/latest/getting_started/v1_user_guide.html
39+
[multimodal]: https://vllm-ascend.readthedocs.io/en/latest/tutorials/single_npu_multimodal.html
40+
[best_of]: https://github.com/vllm-project/vllm/issues/13361
41+
[guided_decoding]: https://github.com/vllm-project/vllm-ascend/issues/177
42+
[v1_scheduler]: https://github.com/vllm-project/vllm/blob/main/vllm/v1/core/sched/scheduler.py
43+
[v1_rfc]: https://github.com/vllm-project/vllm/issues/8779
44+
[multilora]: https://github.com/vllm-project/vllm-ascend/issues/396

0 commit comments

Comments
 (0)