Skip to content

Commit 46ef280

Browse files
menogreymenogrey
authored
[Doc] Add model feature matrix table. (#4040)
### What this PR does / why we need it? Add model feature matrix table. - vLLM version: v0.11.0 - vLLM main: vllm-project/vllm@83f478b Signed-off-by: menogrey <1299267905@qq.com>
1 parent 22286fc commit 46ef280

File tree

1 file changed

+67
-67
lines changed

1 file changed

+67
-67
lines changed

‎docs/source/user_guide/support_matrix/supported_models.md‎

Lines changed: 67 additions & 67 deletions
Original file line numberDiff line numberDiff line change
@@ -6,78 +6,78 @@ Get the latest info here: https://github.com/vllm-project/vllm-ascend/issues/160
66

77
### Generative Models
88

9-
| Model | Support | Note |
10-
|-------------------------------|-----------|----------------------------------------------------------------------|
11-
| DeepSeek V3/3.1 | ✅ | |
12-
| DeepSeek V3.2 EXP | ✅ | |
13-
| DeepSeek R1 | ✅ | |
14-
| DeepSeek Distill (Qwen/LLama) | ✅ | |
15-
| Qwen3 | ✅ | |
16-
| Qwen3-based | ✅ | |
17-
| Qwen3-Coder | ✅ | |
18-
| Qwen3-Moe | ✅ | |
19-
| Qwen3-Next | ✅ | |
20-
| Qwen2.5 | ✅ | |
21-
| Qwen2 | ✅ | |
22-
| Qwen2-based | ✅ | |
23-
| QwQ-32B | ✅ | |
24-
| LLama2/3/3.1 | ✅ | |
25-
| Internlm | ✅ | [#1962](https://github.com/vllm-project/vllm-ascend/issues/1962) |
26-
| Baichuan | ✅ | |
27-
| Baichuan2 | ✅ | |
28-
| Phi-4-mini | ✅ | |
29-
| MiniCPM | ✅ | |
30-
| MiniCPM3 | ✅ | |
31-
| Ernie4.5 | ✅ | |
32-
| Ernie4.5-Moe | ✅ | |
33-
| Gemma-2 | ✅ | |
34-
| Gemma-3 | ✅ | |
35-
| Phi-3/4 | ✅ | |
36-
| Mistral/Mistral-Instruct | ✅ | |
37-
| GLM-4.5 | ✅ | |
38-
| GLM-4 | ❌ | [#2255](https://github.com/vllm-project/vllm-ascend/issues/2255) |
39-
| GLM-4-0414 | ❌ | [#2258](https://github.com/vllm-project/vllm-ascend/issues/2258) |
40-
| ChatGLM | ❌ | [#554](https://github.com/vllm-project/vllm-ascend/issues/554) |
41-
| DeepSeek V2.5 | 🟡 | Need test |
42-
| Mllama | 🟡 | Need test |
43-
| MiniMax-Text | 🟡 | Need test |
9+
| Model | Support | Note | BF16 | Supported Hardware | W8A8 | Chunked Prefill | Automatic Prefix Cache | LoRA | Speculative Decoding | Async Scheduling | Tensor Parallel | Pipeline Parallel | Expert Parallel | Data Parallel | Prefill-decode Disaggregation | Piecewise AclGraph | Fullgraph AclGraph | max-model-len | MLP Weight Prefetch | Doc |
10+
|-------------------------------|-----------|----------------------------------------------------------------------|------|--------------------|------|-----------------|------------------------|------|----------------------|------------------|-----------------|-------------------|-----------------|---------------|-------------------------------|--------------------|--------------------|---------------|---------------------|-----|
11+
| DeepSeek V3/3.1 | ✅ | |||||||||||||||||||
12+
| DeepSeek V3.2 EXP | ✅ | | ✅ | A2/A3 | ✅ | ✅ | ✅ | ✅ | ✅ | | ✅ | ✅ | ✅ | ✅ | ❌ | | | 163840 | | [DeepSeek-V3.2-Exp tutorial](../../tutorials/DeepSeek-V3.2-Exp.md) |
13+
| DeepSeek R1 | ✅ | |||||||||||||||||||
14+
| DeepSeek Distill (Qwen/LLama) | ✅ | |||||||||||||||||||
15+
| Qwen3 | ✅ | |||||||||||||||||||
16+
| Qwen3-based | ✅ | |||||||||||||||||||
17+
| Qwen3-Coder | ✅ | |||||||||||||||||||
18+
| Qwen3-Moe | ✅ | |||||||||||||||||||
19+
| Qwen3-Next | ✅ | |||||||||||||||||||
20+
| Qwen2.5 | ✅ | |||||||||||||||||||
21+
| Qwen2 | ✅ | |||||||||||||||||||
22+
| Qwen2-based | ✅ | |||||||||||||||||||
23+
| QwQ-32B | ✅ | |||||||||||||||||||
24+
| LLama2/3/3.1 | ✅ | |||||||||||||||||||
25+
| Internlm | ✅ | [#1962](https://github.com/vllm-project/vllm-ascend/issues/1962) |||||||||||||||||||
26+
| Baichuan | ✅ | |||||||||||||||||||
27+
| Baichuan2 | ✅ | |||||||||||||||||||
28+
| Phi-4-mini | ✅ | |||||||||||||||||||
29+
| MiniCPM | ✅ | |||||||||||||||||||
30+
| MiniCPM3 | ✅ | |||||||||||||||||||
31+
| Ernie4.5 | ✅ | |||||||||||||||||||
32+
| Ernie4.5-Moe | ✅ | |||||||||||||||||||
33+
| Gemma-2 | ✅ | |||||||||||||||||||
34+
| Gemma-3 | ✅ | |||||||||||||||||||
35+
| Phi-3/4 | ✅ | |||||||||||||||||||
36+
| Mistral/Mistral-Instruct | ✅ | |||||||||||||||||||
37+
| GLM-4.5 | ✅ | |||||||||||||||||||
38+
| GLM-4 | ❌ | [#2255](https://github.com/vllm-project/vllm-ascend/issues/2255) |||||||||||||||||||
39+
| GLM-4-0414 | ❌ | [#2258](https://github.com/vllm-project/vllm-ascend/issues/2258) |||||||||||||||||||
40+
| ChatGLM | ❌ | [#554](https://github.com/vllm-project/vllm-ascend/issues/554) |||||||||||||||||||
41+
| DeepSeek V2.5 | 🟡 | Need test |||||||||||||||||||
42+
| Mllama | 🟡 | Need test |||||||||||||||||||
43+
| MiniMax-Text | 🟡 | Need test |||||||||||||||||||
4444

4545
### Pooling Models
4646

47-
| Model | Support | Note |
48-
|-------------------------------|-----------|----------------------------------------------------------------------|
49-
| Qwen3-Embedding | ✅ | |
50-
| Molmo | ✅ | [1942](https://github.com/vllm-project/vllm-ascend/issues/1942) |
51-
| XLM-RoBERTa-based | ❌ | [1960](https://github.com/vllm-project/vllm-ascend/issues/1960) |
47+
| Model | Support | Note | BF16 | Supported Hardware | W8A8 | Chunked Prefill | Automatic Prefix Cache | LoRA | Speculative Decoding | Async Scheduling | Tensor Parallel | Pipeline Parallel | Expert Parallel | Data Parallel | Prefill-decode Disaggregation | Piecewise AclGraph | Fullgraph AclGraph | max-model-len | MLP Weight Prefetch | Doc |
48+
|-------------------------------|-----------|----------------------------------------------------------------------|------|--------------------|------|-----------------|------------------------|------|----------------------|------------------|-----------------|-------------------|-----------------|---------------|-------------------------------|--------------------|--------------------|---------------|---------------------|-----|
49+
| Qwen3-Embedding | ✅ | |||||||||||||||||||
50+
| Molmo | ✅ | [1942](https://github.com/vllm-project/vllm-ascend/issues/1942) |||||||||||||||||||
51+
| XLM-RoBERTa-based | ❌ | [1960](https://github.com/vllm-project/vllm-ascend/issues/1960) |||||||||||||||||||
5252

5353
## Multimodal Language Models
5454

5555
### Generative Models
5656

57-
| Model | Support | Note |
58-
|--------------------------------|---------------|----------------------------------------------------------------------|
59-
| Qwen2-VL | ✅ | |
60-
| Qwen2.5-VL | ✅ | |
61-
| Qwen3-VL | ✅ | |
62-
| Qwen3-VL-MOE | ✅ | |
63-
| Qwen2.5-Omni | ✅ | [1760](https://github.com/vllm-project/vllm-ascend/issues/1760) |
64-
| QVQ | ✅ | |
65-
| LLaVA 1.5/1.6 | ✅ | [1962](https://github.com/vllm-project/vllm-ascend/issues/1962) |
66-
| InternVL2 | ✅ | |
67-
| InternVL2.5 | ✅ | |
68-
| Qwen2-Audio | ✅ | |
69-
| Aria | ✅ | |
70-
| LLaVA-Next | ✅ | |
71-
| LLaVA-Next-Video | ✅ | |
72-
| MiniCPM-V | ✅ | |
73-
| Mistral3 | ✅ | |
74-
| Phi-3-Vison/Phi-3.5-Vison | ✅ | |
75-
| Gemma3 | ✅ | |
76-
| LLama4 | ❌ | [1972](https://github.com/vllm-project/vllm-ascend/issues/1972) |
77-
| LLama3.2 | ❌ | [1972](https://github.com/vllm-project/vllm-ascend/issues/1972) |
78-
| Keye-VL-8B-Preview | ❌ | [1963](https://github.com/vllm-project/vllm-ascend/issues/1963) |
79-
| Florence-2 | ❌ | [2259](https://github.com/vllm-project/vllm-ascend/issues/2259) |
80-
| GLM-4V | ❌ | [2260](https://github.com/vllm-project/vllm-ascend/issues/2260) |
81-
| InternVL2.0/2.5/3.0<br>InternVideo2.5/Mono-InternVL | ❌ | [2064](https://github.com/vllm-project/vllm-ascend/issues/2064) |
82-
| Whisper | ❌ | [2262](https://github.com/vllm-project/vllm-ascend/issues/2262) |
83-
| Ultravox | 🟡 | Need test |
57+
| Model | Support | Note | BF16 | Supported Hardware | W8A8 | Chunked Prefill | Automatic Prefix Cache | LoRA | Speculative Decoding | Async Scheduling | Tensor Parallel | Pipeline Parallel | Expert Parallel | Data Parallel | Prefill-decode Disaggregation | Piecewise AclGraph | Fullgraph AclGraph | max-model-len | MLP Weight Prefetch | Doc |
58+
|--------------------------------|---------------|----------------------------------------------------------------------|------|--------------------|------|-----------------|------------------------|------|----------------------|------------------|-----------------|-------------------|-----------------|---------------|-------------------------------|--------------------|--------------------|---------------|---------------------|-----|
59+
| Qwen2-VL | ✅ | |||||||||||||||||||
60+
| Qwen2.5-VL | ✅ | |||||||||||||||||||
61+
| Qwen3-VL | ✅ | |||||||||||||||||||
62+
| Qwen3-VL-MOE | ✅ | |||||||||||||||||||
63+
| Qwen2.5-Omni | ✅ | [1760](https://github.com/vllm-project/vllm-ascend/issues/1760) |||||||||||||||||||
64+
| QVQ | ✅ | |||||||||||||||||||
65+
| LLaVA 1.5/1.6 | ✅ | [1962](https://github.com/vllm-project/vllm-ascend/issues/1962) |||||||||||||||||||
66+
| InternVL2 | ✅ | |||||||||||||||||||
67+
| InternVL2.5 | ✅ | |||||||||||||||||||
68+
| Qwen2-Audio | ✅ | |||||||||||||||||||
69+
| Aria | ✅ | |||||||||||||||||||
70+
| LLaVA-Next | ✅ | |||||||||||||||||||
71+
| LLaVA-Next-Video | ✅ | |||||||||||||||||||
72+
| MiniCPM-V | ✅ | |||||||||||||||||||
73+
| Mistral3 | ✅ | |||||||||||||||||||
74+
| Phi-3-Vison/Phi-3.5-Vison | ✅ | |||||||||||||||||||
75+
| Gemma3 | ✅ | |||||||||||||||||||
76+
| LLama4 | ❌ | [1972](https://github.com/vllm-project/vllm-ascend/issues/1972) |||||||||||||||||||
77+
| LLama3.2 | ❌ | [1972](https://github.com/vllm-project/vllm-ascend/issues/1972) |||||||||||||||||||
78+
| Keye-VL-8B-Preview | ❌ | [1963](https://github.com/vllm-project/vllm-ascend/issues/1963) |||||||||||||||||||
79+
| Florence-2 | ❌ | [2259](https://github.com/vllm-project/vllm-ascend/issues/2259) |||||||||||||||||||
80+
| GLM-4V | ❌ | [2260](https://github.com/vllm-project/vllm-ascend/issues/2260) |||||||||||||||||||
81+
| InternVL2.0/2.5/3.0<br>InternVideo2.5/Mono-InternVL | ❌ | [2064](https://github.com/vllm-project/vllm-ascend/issues/2064) |||||||||||||||||||
82+
| Whisper | ❌ | [2262](https://github.com/vllm-project/vllm-ascend/issues/2262) |||||||||||||||||||
83+
| Ultravox | 🟡 | Need test |||||||||||||||||||

0 commit comments

Comments
 (0)