@@ -61,8 +61,8 @@ The llama.cpp CANN backend is designed to support Ascend NPU. It utilize the abi
6161| Llama-2 | √ | √ | √ |
6262| Llama-3 | √ | √ | √ |
6363| Mistral-7B | √ | √ | √ |
64- | Mistral MOE | x | x | x |
65- | DBRX | x | x | x |
64+ | Mistral MOE | √ | √ | √ |
65+ | DBRX | ? | ? | ? |
6666| Falcon | √ | √ | √ |
6767| Chinese LLaMA/Alpaca | √ | √ | √ |
6868| Vigogne(French) | √ | √ | √ |
@@ -80,7 +80,7 @@ The llama.cpp CANN backend is designed to support Ascend NPU. It utilize the abi
8080| Qwen models | √ | √ | √ |
8181| PLaMo-13B | √ | √ | √ |
8282| Phi models | √ | √ | √ |
83- | PhiMoE | x | x | x |
83+ | PhiMoE | ? | ?
10000
| ? |
8484| GPT-2 | √ | √ | √ |
8585| Orion | √ | √ | √ |
8686| InternlLM2 | √ | √ | √ |
@@ -89,45 +89,46 @@ The llama.cpp CANN backend is designed to support Ascend NPU. It utilize the abi
8989| Mamba | √ | √ | √ |
9090| Xverse | √ | √ | √ |
9191| command-r models | √ | √ | √ |
92- | Grok-1 | x | x | x |
92+ | Grok-1 | ? | ? | ? |
9393| SEA-LION | √ | √ | √ |
9494| GritLM-7B | √ | √ | √ |
9595| OLMo | √ | √ | √ |
9696| OLMo 2 | √ | √ | √ |
97- | OLMoE | x | x | x |
97+ | OLMoE | ? | ? | ? |
9898| Granite models | √ | √ | √ |
99- | GPT-NeoX + Pythia | x | x | x |
100- | Snowflake-Arctic MoE | x | × | × |
99+ | GPT-NeoX | ? | ? | ? |
100+ | Pythia | √ | √ | √ |
101+ | Snowflake-Arctic MoE | ? | ? | ? |
101102| Smaug | √ | √ | √ |
102103| Poro 34B | √ | √ | √ |
103- | Bitnet b1.58 models | √ | √ | √ |
104+ | Bitnet b1.58 models | √ | x | x |
104105| Flan-T5 | √ | √ | √ |
105- | Open Elm models | √ | √ | √ |
106+ | Open Elm models | x | x | x |
106107| chatGLM3-6B + ChatGLM4-9b + GLMEdge-1.5b + GLMEdge-4b | √ | √ | √ |
107108| GLM-4-0414 | √ | √ | √ |
108109| SmolLM | √ | √ | √ |
109110| EXAONE-3.0-7.8B-Instruct | √ | √ | √ |
110111| FalconMamba Models | √ | √ | √ |
111- | Jais Models | x | x | x |
112+ | Jais Models | ? | ? | ? |
112113| Bielik-11B-v2.3 | √ | √ | √ |
113- | RWKV-6 | x | x | x |
114- | QRWKV-6 | x | x | x |
114+ | RWKV-6 | √ | √ | √ |
115+ | QRWKV-6 | √ | √ | √ |
115116| GigaChat-20B-A3B | x | x | x |
116117| Trillion-7B-preview | √ | √ | √ |
117118<
CEC3
/td>| Ling models | √ | √ | √ |
118119
119120
120121** Multimodal**
121- | LLaVA 1.5 models, LLaVA 1.6 models | √ | √ | √ |
122- | BakLLaVA | x | x | x |
123- | Obsidian | x | x | x |
124- | ShareGPT4V | x | x | x |
125- | MobileVLM 1.7B/3B models | x | x | x |
126- | Yi-VL | x | x | x |
122+ | LLaVA 1.5 models, LLaVA 1.6 models | ? | ? | ? |
123+ | BakLLaVA | ? | ? | ? |
124+ | Obsidian | ? | ? | ? |
125+ | ShareGPT4V | ? | ? | ? |
126+ | MobileVLM 1.7B/3B models | ? | ? | ? |
127+ | Yi-VL | ? | ? | ? |
127128| Mini CPM | √ | √ | √ |
128129| Moondream | √ | √ | √ |
129- | Bunny | x | x | x |
130- | GLM-EDGE | x | x | x |
130+ | Bunny | ? | ? | ? |
131+ | GLM-EDGE | √ | √ | √ |
131132| Qwen2-VL | √ | √ | √ |
132133
133134
0 commit comments