A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 15, 06:00 UTC
[Model] Add support for MiniMaxM1ForCausalLM (shares architecture with MiniMaxText01ForCausalLM) (#19677)
Signed-off-by: QscQ <qscqesze@gmail.com>
qscqesze committed
Jun 17, 2025 at 00:47 UTC
387bdf0ab98615c1c01fb370ccd9a250b8ac9285
3 files changed
+4
docs/models/supported_models.md
+1
@@ -370,6 +370,7 @@ Specified using `--task generate`.
370
| `TeleChat2ForCausalLM` | TeleChat2 | `Tele-AI/TeleChat2-3B`, `Tele-AI/TeleChat2-7B`, `Tele-AI/TeleChat2-35B`, etc. | ✅︎ | ✅︎ | ✅︎ |
371
| `TeleFLMForCausalLM` | TeleFLM | `CofeAI/FLM-2-52B-Instruct-2407`, `CofeAI/Tele-FLM`, etc. | ✅︎ | ✅︎ | ✅︎ |
372
| `XverseForCausalLM` | XVERSE | `xverse/XVERSE-7B-Chat`, `xverse/XVERSE-13B-Chat`, `xverse/XVERSE-65B-Chat`, etc. | ✅︎ | ✅︎ | ✅︎ |
373
+| `MiniMaxM1ForCausalLM` | MiniMax-Text | `MiniMaxAI/MiniMax-M1-40k`, `MiniMaxAI/MiniMax-M1-80k`etc. | | | |
374
| `MiniMaxText01ForCausalLM` | MiniMax-Text | `MiniMaxAI/MiniMax-Text-01`, etc. | | | |
375
| `Zamba2ForCausalLM` | Zamba2 | `Zyphra/Zamba2-7B-instruct`, `Zyphra/Zamba2-2.7B-instruct`, `Zyphra/Zamba2-1.2B-instruct`, etc. | | | |
376
tests/models/registry.py
+2
@@ -205,6 +205,8 @@ _TEXT_GENERATION_EXAMPLE_MODELS = {
205
trust_remote_code=True),
206
"MiniMaxText01ForCausalLM": _HfExamplesInfo("MiniMaxAI/MiniMax-Text-01",
207
trust_remote_code=True),
208
+ "MiniMaxM1ForCausalLM": _HfExamplesInfo("MiniMaxAI/MiniMax-M1-40k",
209
+ trust_remote_code=True),
210
"MistralForCausalLM": _HfExamplesInfo("mistralai/Mistral-7B-Instruct-v0.1"),
211
"MixtralForCausalLM": _HfExamplesInfo("mistralai/Mixtral-8x7B-Instruct-v0.1", # noqa: E501
212
{"tiny": "TitanML/tiny-mixtral"}), # noqa: E501
vllm/model_executor/models/registry.py
+1
@@ -36,6 +36,7 @@ _TEXT_GENERATION_MODELS = {
36
"AquilaForCausalLM": ("llama", "LlamaForCausalLM"), # AquilaChat2
37
"ArcticForCausalLM": ("arctic", "ArcticForCausalLM"),
38
"MiniMaxText01ForCausalLM": ("minimax_text_01", "MiniMaxText01ForCausalLM"),
39
+ "MiniMaxM1ForCausalLM": ("minimax_text_01", "MiniMaxText01ForCausalLM"),
40
# baichuan-7b, upper case 'C' in the class name
41
"BaiChuanForCausalLM": ("baichuan", "BaiChuanForCausalLM"),
42
# baichuan-13b, lower case 'c' in the class name