Instructions to use MiniMaxAI/MiniMax-M2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use MiniMaxAI/MiniMax-M2 with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="MiniMaxAI/MiniMax-M2", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("MiniMaxAI/MiniMax-M2", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("MiniMaxAI/MiniMax-M2", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Inference
HuggingChat
Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use MiniMaxAI/MiniMax-M2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "MiniMaxAI/MiniMax-M2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "MiniMaxAI/MiniMax-M2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/MiniMaxAI/MiniMax-M2

SGLang

How to use MiniMaxAI/MiniMax-M2 with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "MiniMaxAI/MiniMax-M2" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "MiniMaxAI/MiniMax-M2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "MiniMaxAI/MiniMax-M2" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "MiniMaxAI/MiniMax-M2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use MiniMaxAI/MiniMax-M2 with Docker Model Runner:
```
docker model run hf.co/MiniMaxAI/MiniMax-M2
```

sglang inference error

#46

by birolkuyumcu - opened Dec 10, 2025

Discussion

birolkuyumcu

Dec 10, 2025

docker run --gpus all --rm
-p 30000:8000
--ipc=host
lmsysorg/sglang:dev-cu13
python -m sglang.launch_server
--model-path MiniMaxAI/MiniMax-M2
--tp-size 4
--tool-call-parser minimax-m2
--reasoning-parser minimax-append-think
--host 0.0.0.0
--trust-remote-code
--port 8000
--mem-fraction-static 0.85
this command
with B300 GPU system
rises a

[2025-12-10 05:00:06 TP0] Capture cuda graph bs [1, 2, 4, 8, 12, 16, 24, 32, 40, 48, 56, 64, 72, 80, 88, 96, 104, 112, 120, 128, 136, 144, 152, 160, 168, 176, 184, 192, 200, 208, 216, 224, 232, 240, 248, 256, 272, 288, 304, 320, 336, 352, 368, 384, 400, 416, 432, 448, 464, 480, 496, 512]
[2025-12-10 05:00:06 TP2] Capture cuda graph begin. This can take up to several minutes. avail mem=36.21 GB
Capturing batches (bs=512 avail_mem=35.39 GB): 0%| | 0/52 [00:00<?, ?it/s][2025-12-10 05:00:43 TP2] Using default W8A8 Block FP8 kernel config. Performance might be sub-optimal! Config file not found at /sgl-workspace/sglang/python/sglang/srt/layers/quantization/configs/N=2048,K=3072,device_name=NVIDIA_B300_SXM6_AC,dtype=fp8_w8a8,block_shape=[128, 128].json
[2025-12-10 05:00:46 TP0] Using default W8A8 Block FP8 kernel config. Performance might be sub-optimal! Config file not found at /sgl-workspace/sglang/python/sglang/srt/layers/quantization/configs/N=2048,K=3072,device_name=NVIDIA_B300_SXM6_AC,dtype=fp8_w8a8,block_shape=[128, 128].json
[2025-12-10 05:00:47 TP1] Using default W8A8 Block FP8 kernel config. Performance might be sub-optimal! Config file not found at /sgl-workspace/sglang/python/sglang/srt/layers/quantization/configs/N=2048,K=3072,device_name=NVIDIA_B300_SXM6_AC,dtype=fp8_w8a8,block_shape=[128, 128].json
[2025-12-10 05:00:48 TP3] Using default W8A8 Block FP8 kernel config. Performance might be sub-optimal! Config file not found at /sgl-workspace/sglang/python/sglang/srt/layers/quantization/configs/N=2048,K=3072,device_name=NVIDIA_B300_SXM6_AC,dtype=fp8_w8a8,block_shape=[128, 128].json
[2025-12-10 05:00:49 TP2] Using default W8A8 Block FP8 kernel config. Performance might be sub-optimal! Config file not found at /sgl-workspace/sglang/python/sglang/srt/layers/quantization/configs/N=3072,K=1536,device_name=NVIDIA_B300_SXM6_AC,dtype=fp8_w8a8,block_shape=[128, 128].json
[2025-12-10 05:00:49 TP1] Using default W8A8 Block FP8 kernel config. Performance might be sub-optimal! Config file not found at /sgl-workspace/sglang/python/sglang/srt/layers/quantization/configs/N=3072,K=1536,device_name=NVIDIA_B300_SXM6_AC,dtype=fp8_w8a8,block_shape=[128, 128].json
[2025-12-10 05:00:51 TP3] Using default W8A8 Block FP8 kernel config. Performance might be sub-optimal! Config file not found at /sgl-workspace/sglang/python/sglang/srt/layers/quantization/configs/N=3072,K=1536,device_name=NVIDIA_B300_SXM6_AC,dtype=fp8_w8a8,block_shape=[128, 128].json
[2025-12-10 05:00:52 TP0] Using default W8A8 Block FP8 kernel config. Performance might be sub-optimal! Config file not found at /sgl-workspace/sglang/python/sglang/srt/layers/quantization/configs/N=3072,K=1536,device_name=NVIDIA_B300_SXM6_AC,dtype=fp8_w8a8,block_shape=[128, 128].json
[2025-12-10 05:00:57 TP2] Using default MoE kernel config. Performance might be sub-optimal! Config file not found at /sgl-workspace/sglang/python/sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_5_0/E=256,N=384,device_name=NVIDIA_B300_SXM6_AC,dtype=fp8_w8a8,block_shape=[128, 128].json, you can create them with https://github.com/sgl-project/sglang/tree/main/benchmark/kernels/fused_moe_triton
[2025-12-10 05:00:57 TP2] Using default MoE kernel config. Performance might be sub-optimal! Config file not found at /sgl-workspace/sglang/python/sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_5_0/E=256,N=384,device_name=NVIDIA_B300_SXM6_AC,dtype=fp8_w8a8,block_shape=[128, 128]_down.json, you can create them with https://github.com/sgl-project/sglang/tree/main/benchmark/kernels/fused_moe_triton
[2025-12-10 05:00:59 TP1] Using default MoE kernel config. Performance might be sub-optimal! Config file not found at /sgl-workspace/sglang/python/sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_5_0/E=256,N=384,device_name=NVIDIA_B300_SXM6_AC,dtype=fp8_w8a8,block_shape=[128, 128].json, you can create them with https://github.com/sgl-project/sglang/tree/main/benchmark/kernels/fused_moe_triton
[2025-12-10 05:00:59 TP1] Using default MoE kernel config. Performance might be sub-optimal! Config file not found at /sgl-workspace/sglang/python/sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_5_0/E=256,N=384,device_name=NVIDIA_B300_SXM6_AC,dtype=fp8_w8a8,block_shape=[128, 128]_down.json, you can create them with https://github.com/sgl-project/sglang/tree/main/benchmark/kernels/fused_moe_triton
[2025-12-10 05:00:59 TP3] Using default MoE kernel config. Performance might be sub-optimal! Config file not found at /sgl-workspace/sglang/python/sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_5_0/E=256,N=384,device_name=NVIDIA_B300_SXM6_AC,dtype=fp8_w8a8,block_shape=[128, 128].json, you can create them with https://github.com/sgl-project/sglang/tree/main/benchmark/kernels/fused_moe_triton
[2025-12-10 05:00:59 TP3] Using default MoE kernel config. Performance might be sub-optimal! Config file not found at /sgl-workspace/sglang/python/sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_5_0/E=256,N=384,device_name=NVIDIA_B300_SXM6_AC,dtype=fp8_w8a8,block_shape=[128, 128]_down.json, you can create them with https://github.com/sgl-project/sglang/tree/main/benchmark/kernels/fused_moe_triton
[2025-12-10 05:01:00 TP0] Using default MoE kernel config. Performance might be sub-optimal! Config file not found at /sgl-workspace/sglang/python/sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_5_0/E=256,N=384,device_name=NVIDIA_B300_SXM6_AC,dtype=fp8_w8a8,block_shape=[128, 128].json, you can create them with https://github.com/sgl-project/sglang/tree/main/benchmark/kernels/fused_moe_triton
[2025-12-10 05:01:00 TP0] Using default MoE kernel config. Performance might be sub-optimal! Config file not found at /sgl-workspace/sglang/python/sglang/srt/layers/moe/fused_moe_triton/configs/triton_3_5_0/E=256,N=384,device_name=NVIDIA_B300_SXM6_AC,dtype=fp8_w8a8,block_shape=[128, 128]_down.json, you can create them with https://github.com/sgl-project/sglang/tree/main/benchmark/kernels/fused_moe_triton
Capturing batches (bs=2 avail_mem=33.94 GB): 96%|█████████▌| 50/52 [01:21<00:03, 1.63s/it]
[2025-12-10 05:01:28 TP0] Registering 12450 cuda graph addresses
[2025-12-10 05:01:29 TP3] Scheduler hit an exception: Traceback (most recent call last):
File "/sgl-workspace/sglang/python/sglang/srt/managers/scheduler.py", line 2703, in run_scheduler_process
scheduler = Scheduler(
^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/managers/scheduler.py", line 312, in init
self.tp_worker = TpModelWorker(
^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/managers/tp_worker.py", line 237, in init
self._model_runner = ModelRunner(
^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/model_executor/model_runner.py", line 323, in init
self.initialize(min_per_gpu_memory)
File "/sgl-workspace/sglang/python/sglang/srt/model_executor/model_runner.py", line 490, in initialize
self.init_device_graphs()
File "/sgl-workspace/sglang/python/sglang/srt/model_executor/model_runner.py", line 2006, in init_device_graphs
self.graph_runner = graph_runnersself.device
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/model_executor/cuda_graph_runner.py", line 390, in init
self.capture()
File "/sgl-workspace/sglang/python/sglang/srt/model_executor/cuda_graph_runner.py", line 548, in capture
_capture_one_stream()
File "/sgl-workspace/sglang/python/sglang/srt/model_executor/cuda_graph_runner.py", line 532, in _capture_one_stream
) = self.capture_one_batch_size(bs, forward, stream_idx)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/model_executor/cuda_graph_runner.py", line 738, in capture_one_batch_size
run_once()
File "/sgl-workspace/sglang/python/sglang/srt/model_executor/cuda_graph_runner.py", line 725, in run_once
logits_output_or_pp_proxy_tensors = forward(
^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/utils/_contextlib.py", line 120, in decorate_context
return func(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/models/minimax_m2.py", line 833, in forward
hidden_states = self.model(input_ids, positions, forward_batch, input_embeds)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/nn/modules/module.py", line 1775, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/nn/modules/module.py", line 1786, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/models/minimax_m2.py", line 751, in forward
hidden_states, residual = layer(
^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/nn/modules/module.py", line 1775, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/nn/modules/module.py", line 1786, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/models/minimax_m2.py", line 588, in forward
hidden_states = self.self_attn(
^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/nn/modules/module.py", line 1775, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/nn/modules/module.py", line 1786, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/models/minimax_m2.py", line 503, in forward
s = self.forward_prepare(
^^^^^^^^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/models/minimax_m2.py", line 483, in forward_prepare
q = self.q_norm(q.contiguous())
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/nn/modules/module.py", line 1775, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/nn/modules/module.py", line 1786, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/_dynamo/eval_frame.py", line 832, in compile_wrapper
return fn(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/models/minimax_m2.py", line 104, in forward
@torch .compile(dynamic=True, backend=get_compiler_backend())
File "/usr/local/lib/python3.12/dist-packages/torch/_dynamo/eval_frame.py", line 1044, in _fn
return fn(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/_functorch/aot_autograd.py", line 1130, in forward
return compiled_fn(full_args)
^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/_functorch/_aot_autograd/runtime_wrappers.py", line 353, in runtime_wrapper
all_outs = call_func_at_runtime_with_args(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/_functorch/_aot_autograd/utils.py", line 129, in call_func_at_runtime_with_args
out = normalize_as_list(f(args))
^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/_functorch/_aot_autograd/runtime_wrappers.py", line 724, in inner_fn
outs = compiled_fn(args)
^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/_functorch/_aot_autograd/runtime_wrappers.py", line 526, in wrapper
return compiled_fn(runtime_args)
^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/_inductor/output_code.py", line 613, in call
return self.current_callable(inputs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/_inductor/utils.py", line 2962, in run
out = model(new_inputs)
^^^^^^^^^^^^^^^^^
File "/tmp/torchinductor_root/3y/c3y5nx5cuasmyf7sxjae2saikpdj5vg6qhmbfksz5lguvrkz47lb.py", line 193, in call
buf2 = torch.ops.sglang.outplace_all_reduce.default(buf1, 'tp:0', 'ca')
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/_ops.py", line 841, in call
return self._op(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py", line 156, in outplace_all_reduce
return group._all_reduce_out_place(tensor, outplace_all_reduce_method)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py", line 645, in _all_reduce_out_place
assert out is not None
^^^^^^^^^^^^^^^
AssertionError

[2025-12-10 05:01:29] Received sigquit from a child process. It usually means the child failed.
[2025-12-10 05:01:29 TP1] Scheduler hit an exception: Traceback (most recent call last):
File "/sgl-workspace/sglang/python/sglang/srt/managers/scheduler.py", line 2703, in run_scheduler_process
scheduler = Scheduler(
^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/managers/scheduler.py", line 312, in init
self.tp_worker = TpModelWorker(
^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/managers/tp_worker.py", line 237, in init
self._model_runner = ModelRunner(
^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/model_executor/model_runner.py", line 323, in init
self.initialize(min_per_gpu_memory)
File "/sgl-workspace/sglang/python/sglang/srt/model_executor/model_runner.py", line 490, in initialize
self.init_device_graphs()
File "/sgl-workspace/sglang/python/sglang/srt/model_executor/model_runner.py", line 2006, in init_device_graphs
self.graph_runner = graph_runnersself.device
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/model_executor/cuda_graph_runner.py", line 390, in init
self.capture()
File "/sgl-workspace/sglang/python/sglang/srt/model_executor/cuda_graph_runner.py", line 548, in capture
_capture_one_stream()
File "/sgl-workspace/sglang/python/sglang/srt/model_executor/cuda_graph_runner.py", line 532, in _capture_one_stream
) = self.capture_one_batch_size(bs, forward, stream_idx)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/model_executor/cuda_graph_runner.py", line 738, in capture_one_batch_size
run_once()
File "/sgl-workspace/sglang/python/sglang/srt/model_executor/cuda_graph_runner.py", line 725, in run_once
logits_output_or_pp_proxy_tensors = forward(
^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/utils/_contextlib.py", line 120, in decorate_context
return func(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/models/minimax_m2.py", line 833, in forward
hidden_states = self.model(input_ids, positions, forward_batch, input_embeds)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/nn/modules/module.py", line 1775, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/nn/modules/module.py", line 1786, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/models/minimax_m2.py", line 751, in forward
hidden_states, residual = layer(
^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/nn/modules/module.py", line 1775, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/nn/modules/module.py", line 1786, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/models/minimax_m2.py", line 588, in forward
hidden_states = self.self_attn(
^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/nn/modules/module.py", line 1775, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/nn/modules/module.py", line 1786, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/models/minimax_m2.py", line 503, in forward
s = self.forward_prepare(
^^^^^^^^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/models/minimax_m2.py", line 483, in forward_prepare
q = self.q_norm(q.contiguous())
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/nn/modules/module.py", line 1775, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/nn/modules/module.py", line 1786, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/_dynamo/eval_frame.py", line 832, in compile_wrapper
return fn(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/models/minimax_m2.py", line 104, in forward
@torch .compile(dynamic=True, backend=get_compiler_backend())
File "/usr/local/lib/python3.12/dist-packages/torch/_dynamo/eval_frame.py", line 1044, in _fn
return fn(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/_functorch/aot_autograd.py", line 1130, in forward
return compiled_fn(full_args)
^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/_functorch/_aot_autograd/runtime_wrappers.py", line 353, in runtime_wrapper
all_outs = call_func_at_runtime_with_args(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/_functorch/_aot_autograd/utils.py", line 129, in call_func_at_runtime_with_args
out = normalize_as_list(f(args))
^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/_functorch/_aot_autograd/runtime_wrappers.py", line 724, in inner_fn
outs = compiled_fn(args)
^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/_functorch/_aot_autograd/runtime_wrappers.py", line 526, in wrapper
return compiled_fn(runtime_args)
^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/_inductor/output_code.py", line 613, in call
return self.current_callable(inputs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/_inductor/utils.py", line 2962, in run
out = model(new_inputs)
^^^^^^^^^^^^^^^^^
File "/tmp/torchinductor_root/xs/cxszet6rwdjswaamvzjzakal5gwfkyn4csisn55tgpef2z7twnrl.py", line 193, in call
buf2 = torch.ops.sglang.outplace_all_reduce.default(buf1, 'tp:0', 'ca')
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/_ops.py", line 841, in call
return self._op(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py", line 156, in outplace_all_reduce
return group._all_reduce_out_place(tensor, outplace_all_reduce_method)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py", line 645, in _all_reduce_out_place
assert out is not None
^^^^^^^^^^^^^^^
AssertionError

[2025-12-10 05:01:29] Received sigquit from a child process. It usually means the child failed.
[2025-12-10 05:01:29 TP2] Scheduler hit an exception: Traceback (most recent call last):
File "/sgl-workspace/sglang/python/sglang/srt/managers/scheduler.py", line 2703, in run_scheduler_process
scheduler = Scheduler(
^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/managers/scheduler.py", line 312, in init
self.tp_worker = TpModelWorker(
^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/managers/tp_worker.py", line 237, in init
self._model_runner = ModelRunner(
^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/model_executor/model_runner.py", line 323, in init
self.initialize(min_per_gpu_memory)
File "/sgl-workspace/sglang/python/sglang/srt/model_executor/model_runner.py", line 490, in initialize
self.init_device_graphs()
File "/sgl-workspace/sglang/python/sglang/srt/model_executor/model_runner.py", line 2006, in init_device_graphs
self.graph_runner = graph_runnersself.device
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/model_executor/cuda_graph_runner.py", line 390, in init
self.capture()
File "/sgl-workspace/sglang/python/sglang/srt/model_executor/cuda_graph_runner.py", line 548, in capture
_capture_one_stream()
File "/sgl-workspace/sglang/python/sglang/srt/model_executor/cuda_graph_runner.py", line 532, in _capture_one_stream
) = self.capture_one_batch_size(bs, forward, stream_idx)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/model_executor/cuda_graph_runner.py", line 738, in capture_one_batch_size
run_once()
File "/sgl-workspace/sglang/python/sglang/srt/model_executor/cuda_graph_runner.py", line 725, in run_once
logits_output_or_pp_proxy_tensors = forward(
^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/utils/_contextlib.py", line 120, in decorate_context
return func(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/models/minimax_m2.py", line 833, in forward
hidden_states = self.model(input_ids, positions, forward_batch, input_embeds)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/nn/modules/module.py", line 1775, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/nn/modules/module.py", line 1786, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/models/minimax_m2.py", line 751, in forward
hidden_states, residual = layer(
^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/nn/modules/module.py", line 1775, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/nn/modules/module.py", line 1786, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/models/minimax_m2.py", line 588, in forward
hidden_states = self.self_attn(
^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/nn/modules/module.py", line 1775, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/nn/modules/module.py", line 1786, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/models/minimax_m2.py", line 503, in forward
s = self.forward_prepare(
^^^^^^^^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/models/minimax_m2.py", line 483, in forward_prepare
q = self.q_norm(q.contiguous())
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/nn/modules/module.py", line 1775, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/nn/modules/module.py", line 1786, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/_dynamo/eval_frame.py", line 832, in compile_wrapper
return fn(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/models/minimax_m2.py", line 104, in forward
@torch .compile(dynamic=True, backend=get_compiler_backend())
File "/usr/local/lib/python3.12/dist-packages/torch/_dynamo/eval_frame.py", line 1044, in _fn
return fn(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/_functorch/aot_autograd.py", line 1130, in forward
return compiled_fn(full_args)
^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/_functorch/_aot_autograd/runtime_wrappers.py", line 353, in runtime_wrapper
all_outs = call_func_at_runtime_with_args(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/_functorch/_aot_autograd/utils.py", line 129, in call_func_at_runtime_with_args
out = normalize_as_list(f(args))
^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/_functorch/_aot_autograd/runtime_wrappers.py", line 724, in inner_fn
outs = compiled_fn(args)
^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/_functorch/_aot_autograd/runtime_wrappers.py", line 526, in wrapper
return compiled_fn(runtime_args)
^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/_inductor/output_code.py", line 613, in call
return self.current_callable(inputs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/_inductor/utils.py", line 2962, in run
out = model(new_inputs)
^^^^^^^^^^^^^^^^^
File "/tmp/torchinductor_root/7x/c7x6a6phuc3hl2dwxu2uazp53qjt3432smmyovpmmtokvxq42mjt.py", line 193, in call
buf2 = torch.ops.sglang.outplace_all_reduce.default(buf1, 'tp:0', 'ca')
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/_ops.py", line 841, in call
return self._op(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py", line 156, in outplace_all_reduce
return group._all_reduce_out_place(tensor, outplace_all_reduce_method)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py", line 645, in _all_reduce_out_place
assert out is not None
^^^^^^^^^^^^^^^
AssertionError

[2025-12-10 05:01:29] Received sigquit from a child process. It usually means the child failed.
[2025-12-10 05:01:29 TP0] Scheduler hit an exception: Traceback (most recent call last):
File "/sgl-workspace/sglang/python/sglang/srt/managers/scheduler.py", line 2703, in run_scheduler_process
scheduler = Scheduler(
^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/managers/scheduler.py", line 312, in init
self.tp_worker = TpModelWorker(
^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/managers/tp_worker.py", line 237, in init
self._model_runner = ModelRunner(
^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/model_executor/model_runner.py", line 323, in init
self.initialize(min_per_gpu_memory)
File "/sgl-workspace/sglang/python/sglang/srt/model_executor/model_runner.py", line 490, in initialize
self.init_device_graphs()
File "/sgl-workspace/sglang/python/sglang/srt/model_executor/model_runner.py", line 2006, in init_device_graphs
self.graph_runner = graph_runnersself.device
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/model_executor/cuda_graph_runner.py", line 390, in init
self.capture()
File "/sgl-workspace/sglang/python/sglang/srt/model_executor/cuda_graph_runner.py", line 548, in capture
_capture_one_stream()
File "/sgl-workspace/sglang/python/sglang/srt/model_executor/cuda_graph_runner.py", line 532, in _capture_one_stream
) = self.capture_one_batch_size(bs, forward, stream_idx)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/model_executor/cuda_graph_runner.py", line 738, in capture_one_batch_size
run_once()
File "/sgl-workspace/sglang/python/sglang/srt/model_executor/cuda_graph_runner.py", line 725, in run_once
logits_output_or_pp_proxy_tensors = forward(
^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/utils/_contextlib.py", line 120, in decorate_context
return func(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/models/minimax_m2.py", line 833, in forward
hidden_states = self.model(input_ids, positions, forward_batch, input_embeds)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/nn/modules/module.py", line 1775, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/nn/modules/module.py", line 1786, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/models/minimax_m2.py", line 751, in forward
hidden_states, residual = layer(
^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/nn/modules/module.py", line 1775, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/nn/modules/module.py", line 1786, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/models/minimax_m2.py", line 588, in forward
hidden_states = self.self_attn(
^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/nn/modules/module.py", line 1775, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/nn/modules/module.py", line 1786, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/models/minimax_m2.py", line 503, in forward
s = self.forward_prepare(
^^^^^^^^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/models/minimax_m2.py", line 483, in forward_prepare
q = self.q_norm(q.contiguous())
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/nn/modules/module.py", line 1775, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/nn/modules/module.py", line 1786, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/_dynamo/eval_frame.py", line 832, in compile_wrapper
return fn(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/models/minimax_m2.py", line 104, in forward
@torch .compile(dynamic=True, backend=get_compiler_backend())
File "/usr/local/lib/python3.12/dist-packages/torch/_dynamo/eval_frame.py", line 1044, in _fn
return fn(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/_functorch/aot_autograd.py", line 1130, in forward
return compiled_fn(full_args)
^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/_functorch/_aot_autograd/runtime_wrappers.py", line 353, in runtime_wrapper
all_outs = call_func_at_runtime_with_args(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/_functorch/_aot_autograd/utils.py", line 129, in call_func_at_runtime_with_args
out = normalize_as_list(f(args))
^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/_functorch/_aot_autograd/runtime_wrappers.py", line 724, in inner_fn
outs = compiled_fn(args)
^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/_functorch/_aot_autograd/runtime_wrappers.py", line 526, in wrapper
return compiled_fn(runtime_args)
^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/_inductor/output_code.py", line 613, in call
return self.current_callable(inputs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/_inductor/utils.py", line 2962, in run
out = model(new_inputs)
^^^^^^^^^^^^^^^^^
File "/tmp/torchinductor_root/ht/chtdfqxs7kuh6gjaahminfcvejtw22jfhd4bxditf4tmpa65rdi6.py", line 193, in call
buf2 = torch.ops.sglang.outplace_all_reduce.default(buf1, 'tp:0', 'ca')
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/dist-packages/torch/_ops.py", line 841, in call
return self._op(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py", line 156, in outplace_all_reduce
return group._all_reduce_out_place(tensor, outplace_all_reduce_method)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/distributed/parallel_state.py", line 645, in _all_reduce_out_place
assert out is not None
^^^^^^^^^^^^^^^
AssertionError

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment