qwen3.8-max#

  • Context Length: 262144

  • Model Name: qwen3.8-max

  • Languages: en, zh

  • Abilities: chat, tools, reasoning

  • Description: Qwen3.8-2.4T-A95B is a text-only Mixture-of-Experts model with 2.4 trillion total parameters and 95 billion activated parameters, designed for coding, research, professional work, and long-horizon agentic tasks.

Specifications#

Model Spec 1 (pytorch, 2400 Billion)#

  • Model Format: pytorch

  • Model Size (in billions): 2400

  • Quantizations: none

  • Engines: vLLM, Transformers, SGLang

  • Model ID: Qwen/Qwen3.8-2.4T-A95B

  • Model Hubs: Hugging Face, ModelScope

Execute the following command to launch the model, remember to replace ${quantization} with your chosen quantization method from the options listed above:

xinference launch --model-engine ${engine} --model-name qwen3.8-max --size-in-billions 2400 --model-format pytorch --quantization ${quantization}

Model Spec 2 (fp8, 2400 Billion)#

  • Model Format: fp8

  • Model Size (in billions): 2400

  • Quantizations: FP8

  • Engines: vLLM, SGLang

  • Model ID: Qwen/Qwen3.8-2.4T-A95B-FP8

  • Model Hubs: Hugging Face, ModelScope

Execute the following command to launch the model, remember to replace ${quantization} with your chosen quantization method from the options listed above:

xinference launch --model-engine ${engine} --model-name qwen3.8-max --size-in-billions 2400 --model-format fp8 --quantization ${quantization}

Model Spec 3 (ggufv2, 2400 Billion)#

  • Model Format: ggufv2

  • Model Size (in billions): 2400

  • Quantizations: BF16, Q8_0, UD-IQ1_M, UD-IQ1_S, UD-IQ2_XS, UD-IQ2_XXS, UD-IQ3_XXS, UD-IQ4_XS, UD-Q1_0

  • Engines: vLLM, llama.cpp

  • Model ID: unsloth/Qwen3.8-2.4T-A95B-GGUF

  • Model Hubs: Hugging Face, ModelScope

Execute the following command to launch the model, remember to replace ${quantization} with your chosen quantization method from the options listed above:

xinference launch --model-engine ${engine} --model-name qwen3.8-max --size-in-billions 2400 --model-format ggufv2 --quantization ${quantization}