MindSpeed-LLM MindSpore后端稠密模型支持详情
MindSpore后端仅支持以mcore方式实现的稠密模型,对Legacy方式实现的模型无支持计划。
具体支持情况见下表,持续更新中。
| 模型 | 下载链接 | 脚本位置 | 序列 | 实现 | 集群 | 是否支持 |
|---|---|---|---|---|---|---|
| ChatGLM3 | 6B | chatglm3 | 8K | Mcore | 1x8 | ✅ |
| 64K | Mcore | 2x8 | ✅ | |||
| GLM4 | 9B | glm4 | 8K | Mcore | 1x8 | ✅ |
| 32K | Mcore | 2x8 | ✅ | |||
| 32B | 8K | Mcore | 2x8 | ✅ | ||
| CodeLlama | 34B | codellama | 4K | Mcore | 2x8 | ✅ |
| InternLM2 | 20B | internlm2 | 4K | Mcore | 1x8 | ✅ |
| 32K | Mcore | 1x8 | ✅ | |||
| InternLM2.5 | 1.8B | internlm25 | 32K | Mcore | 1x8 | ✅ |
| 7B | 32K | Mcore | 1x8 | ✅ | ||
| 20B | 32K | Mcore | 2x8 | ✅ | ||
| LLaMA2 | 7B | llama2 | 4K | Mcore | 1x8 | ✅ |
| 16K | Mcore | 1x8 | ✅ | |||
| 32K | Mcore | 1x8 | ✅ | |||
| 13B | 4K | Mcore | 1x8 | ✅ | ||
| 34B | 4K | Mcore | 2x8 | ✅ | ||
| 70B | 4K | Mcore | 4x8 | ✅ | ||
| LLaMA3 | 8B | llama3 | 8K | Mcore | 1x8 | ✅ |
| 70B | 8K | Mcore | 4x8 | ✅ | ||
| LLaMA3.1 | 8B | llama31 | 8K | Mcore | 1x8 | ✅ |
| 128K | Mcore | 4x8 | 支持中 | |||
| 70B | 8K | Mcore | 4x8 | ✅ | ||
| LLaMA3.2 | 1B | llama32 | 8K | Mcore | 1x8 | ✅ |
| 3B | 8K | Mcore | 1x8 | ✅ | ||
| LLaMA3.3 | 70B-Instruct | llama33 | 8K | Mcore | 4x8 | ✅ |
| Qwen1.5 | 0.5B | qwen15 | 8K | Mcore | 1x8 | ✅ |
| 1.8B | 8K | Mcore | 1x8 | ✅ | ||
| 4B | 8K | Mcore | 1x8 | ✅ | ||
| 7B | 8K | Mcore | 1x8 | ✅ | ||
| 14B | 8K | Mcore | 1x8 | ✅ | ||
| 32B | 8K | Mcore | 4x8 | ✅ | ||
| 72B | 8K | Mcore | 8x8 | ✅ | ||
| 110B | 8K | Mcore | 8x8 | ✅ | ||
| CodeQwen1.5 | 7B | 8K | Mcore | 1x8 | ✅ | |
| Qwen2 | 0.5B | qwen2 | 4K | Mcore | 1x8 | ✅ |
| 32K | Mcore | 1x8 | ✅ | |||
| 1.5B | 4K | Mcore | 1x8 | ✅ | ||
| 32K | Mcore | 1x8 | ✅ | |||
| 7B | 4K | Mcore | 1x8 | ✅ | ||
| 32K | Mcore | 1x8 | ✅ | |||
| 72B | 4K | Mcore | 4x8 | ✅ | ||
| Qwen2.5 | 0.5B | qwen25 | 32K | Mcore | 1x8 | ✅ | 1.5B | 32K | Mcore | 1x8 | ✅ | 3B | 32K | Mcore | 1x8 | ✅ |
| 7B | 4K | Mcore | 1x8 | ✅ | ||
| 32K | Mcore | 1x8 | ✅ | |||
| 14B | 4K | Mcore | 1x8 | ✅ | 32K | Mcore | 2x8 | ✅ |
| 32B | 4K | Mcore | 4x8 | ✅ | ||
| 32K | Mcore | 4x8 | ✅ | |||
| 72B | 4K | Mcore | 4x8 | ✅ | ||
| 32K | Mcore | 8x8 | ✅ | |||
| Qwen3 | 0.6B | qwen3 | 4K | Mcore | 1x8 | ✅ |
| 1.7B | 4K | Mcore | 1x8 | ✅ | ||
| 4B | 4K | Mcore | 1x8 | ✅ | ||
| 8B | 4K | Mcore | 1x8 | ✅ | ||
| 14B | 4K | Mcore | 1x8 | ✅ | ||
| 32B | 4K | Mcore | 2x8 | ✅ | ||
| Qwen2.5-Math | 1.5B | qwen25_math | 4K | Mcore | 1x8 | ✅ |
| 7B | 4K | Mcore | 1x8 | ✅ | ||
| 72B | 4K | Mcore | 4x8 | ✅ | ||
| Yi | 34B | yi | 4K | Mcore | 2x8 | ✅ |
| Yi1.5 | 6B | yi15 | 4K | Mcore | 1x8 | ✅ |
| 34B | 4K | Mcore | 2x8 | ✅ | ||
| Mistral | 7B | mistral | 32K | Mcore | 1x8 | ✅ |
| Gemma | 2B | gemma | 8K | Mcore | 1x8 | ✅ |
| 7B | 8K | Mcore | 1x8 | ✅ | ||
| Gemma2 | 9B | gemma2 | 8K | Mcore | 1x8 | ✅ |
| 27B | 8K | Mcore | 2x8 | ✅ | ||
| grok-1 | 40B | grok-1 | 8K | Mcore | 4x8 | ✅ |
| MiniCPM | 2B | minicpm | 4K | Mcore | 1x8 | ✅ |
| Phi3.5 | mini-instruct | phi35 | 4K | Mcore | 1x8 | ✅ |
| DeepSeek-R1-Distill-Qwen | 1.5B | deepseek_r1_distill_qwen | 4K | Mcore | 1x8 | ✅ |
| 7B | 4K | Mcore | 1x8 | ✅ | ||
| 14B | 4K | Mcore | 1x8 | ✅ | ||
| 32B | 8K | Mcore | 2x8 | ✅ | ||
| DeepSeek-R1-Distill-LLaMA | 8B | deepseek_r1_distill_llama | 8K | Mcore | 1x8 | ✅ |
| 70B | 8K | Mcore | 4x8 | ✅ |