inference-optimization/Qwen3-235B-A22B-Thinking-2507-quantized.w8a8
Qwen3-235B-A22B-Thinking-2507-quantized.w8a8
Model Overview
- Model Architecture: Qwen3ForCausalLM
- Input: Text
- Output: Text
- Model Optimizations:
- Weight quantization: INT8
- Intended Use Cases:
- Reasoning.
- Function calling.
- Subject matter experts via fine-tuning.
- Multilingual instruction following.
- Translation.
- Out-of-scope: Use in any manner that violates applicable laws or regulations (including trade compliance laws).
- Release Date: 15/05/2025
- Version: 1.0
- Model Developers: RedHat (Neural Magic)
Model Optimizations
This model was obtained by quantizing the weights of Qwen/Qwen3-235B-A22B-Thinking-2507 to INT8 data type. This optimization reduces the number of bits used to represent weights and activations from 16 to 8, reducing GPU memory requirements (by approximately 50%) and increasing matrix-multiply compute throughput (by approximately 2x). Weight quantization also reduces disk size requirements by approximately 50%.
Only weights and activations of the linear operators within transformers blocks are quantized. Weights are quantized with a symmetric static per-channel scheme, whereas activations are quantized with a symmetric dynamic per-token scheme. A combination of the SmoothQuant and GPTQ algorithms is applied for quantization, as implemented in the llm-compressor library.
Deployment
This model can be deployed efficiently using the vLLM backend, as shown in the example below.
from vllm import LLM, SamplingParams
from transformers import AutoTokenizer
model_id = "RedHatAI/Qwen3-235B-A22B-Thinking-2507-quantized.w8a8"
number_gpus = 1
sampling_params = SamplingParams(temperature=0.6, top_p=0.95, top_k=20, min_p=0, max_tokens=256)
messages = [
{"role": "user", "content": prompt}
]
tokenizer = AutoTokenizer.from_pretrained(model_id)
messages = [{"role": "user", "content": "Give me a short introduction to large language model."}]
prompts = tokenizer.apply_chat_template(messages, add_generation_prompt=True, tokenize=False)
llm = LLM(model=model_id, tensor_parallel_size=number_gpus)
outputs = llm.generate(prompts, sampling_params)
generated_text = outputs[0].outputs[0].text
print(generated_text)vLLM aslo supports OpenAI-compatible serving. See the documentation for more details.
Creation
<details> <summary>Creation details</summary> This model was created with llm-compressor by running the code snippet below.
from llmcompressor.modifiers.quantization import GPTQModifier
from llmcompressor.modifiers.smoothquant import SmoothQuantModifier
from llmcompressor.transformers import oneshot
from transformers import AutoModelForCausalLM, AutoTokenizer
# Load model
model_stub = "Qwen/Qwen3-30B-A3B-Thinking"
model_name = model_stub.split("/")[-1]
num_samples = 1024
max_seq_len = 8192
model = AutoModelForCausalLM.from_pretrained(model_stub)
tokenizer = AutoTokenizer.from_pretrained(model_stub)
def preprocess_fn(example):
return {"text": tokenizer.apply_chat_template(example["messages"], add_generation_prompt=False, tokenize=False)}
ds = load_dataset("neuralmagic/LLM_compression_calibration", split="train")
ds = ds.map(preprocess_fn)
# Configure the quantization algorithm and scheme
recipe = [
SmoothQuantModifier(
smoothing_strength=0.9,
mappings=[
[["re:.*q_proj", "re:.*k_proj", "re:.*v_proj"], "re:.*input_layernorm"],
],
),
GPTQModifier(
ignore: ["lm_head"]
config_groups={"group_0": {"targets": ["Linear"], "weights": { "num_bits": 4, "type": int, "strategy": "group", "group_size": 128, "symmetric": true, "dynamic": false, "observer": "mse" } } },
dampening_frac=0.1,
)
]
# Apply quantization
oneshot(
model=model,
dataset=ds,
recipe=recipe,
max_seq_length=max_seq_len,
num_calibration_samples=num_samples,
)
# Save to disk in compressed-tensors format
save_path = model_name + "-quantized.w8a8"
model.save_pretrained(save_path)
tokenizer.save_pretrained(save_path)
print(f"Model and tokenizer saved to: {save_path}")</details>
Evaluation
The model was evaluated on the ifeval, mmlupro and gsm8kplatinum using lm-evaluation-harness, on reasoning tasks using lighteval. vLLM was used for all evaluations.
<details> <summary>Evaluation details</summary>
Deploy using vllm to create an OpenAI-compatible API endpoint:
- vLLM:
vllm serve RedHatAI/Qwen3-235B-A22B-Thinking-2507-quantized.w8a8 --max-model-len 262144 --reasoning-parser deepseek_r1 --tensor-parallel-size 8lm-evaluation-harness
lm_eval --model local-chat-completions \
--tasks mmlu_pro_chat \
--model_args "model=RedHatAI/Qwen3-235B-A22B-Thinking-2507-quantized.w8a8,max_length=262144,base_url=http://0.0.0.0:8000/v1/chat/completions,num_concurrent=64,max_retries=3,tokenized_requests=False,tokenizer_backend=None,timeout=1200" \
--num_fewshot 0 \
--apply_chat_template \
--gen_kwargs "do_sample=True,temperature=0.6,top_p=0.95,top_k=20,min_p=0.0,max_gen_toks=64000 lm_eval --model local-chat-completions \
--tasks ifeval \
--model_args "model=RedHatAI/Qwen3-235B-A22B-Thinking-2507-quantized.w8a8,max_length=262144,base_url=http://0.0.0.0:8000/v1/chat/completions,num_concurrent=64,max_retries=3,tokenized_requests=False,tokenizer_backend=None,timeout=1200" \
--num_fewshot 0 \
--apply_chat_template \
--gen_kwargs "do_sample=True,temperature=0.6,top_p=0.95,top_k=20,min_p=0.0,max_gen_toks=64000 lm_eval --model local-chat-completions \
--tasks gsm8k_platinum_cot_llama \
--model_args "model=RedHatAI/Qwen3-235B-A22B-Thinking-2507-quantized.w8a8,max_length=262144,base_url=http://0.0.0.0:8000/v1/chat/completions,num_concurrent=64,max_retries=3,tokenized_requests=False,tokenizer_backend=None,timeout=1200" \
--num_fewshot 0 \
--apply_chat_template \
--gen_kwargs "do_sample=True,temperature=0.6,top_p=0.95,top_k=20,min_p=0.0,max_gen_toks=64000lighteval
lightevalmodelarguments.yaml
model_parameters:
model_name: RedHatAI/Qwen3-235B-A22B-Thinking-2507-quantized.w8a8
dtype: auto
gpu_memory_utilization: 0.9
max_model_length: 40960
generation_parameters:
temperature: 0.6
top_k: 20
min_p: 0.0
top_p: 0.95
max_new_tokens: 32000 lighteval endpoint litellm lighteval_model_arguments.yaml \
"aime25|0,math_500|0,gpqa:diamond|0"</details>