Instructions to use willamazon1/qwen3-8b-tmax-sft-v3-iter353 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use willamazon1/qwen3-8b-tmax-sft-v3-iter353 with Transformers:

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

pipe = pipeline("text-generation", model="willamazon1/qwen3-8b-tmax-sft-v3-iter353")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("willamazon1/qwen3-8b-tmax-sft-v3-iter353")
model = AutoModelForCausalLM.from_pretrained("willamazon1/qwen3-8b-tmax-sft-v3-iter353", device_map="auto")
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]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use willamazon1/qwen3-8b-tmax-sft-v3-iter353 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "willamazon1/qwen3-8b-tmax-sft-v3-iter353"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "willamazon1/qwen3-8b-tmax-sft-v3-iter353",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/willamazon1/qwen3-8b-tmax-sft-v3-iter353

SGLang

How to use willamazon1/qwen3-8b-tmax-sft-v3-iter353 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 "willamazon1/qwen3-8b-tmax-sft-v3-iter353" \
    --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": "willamazon1/qwen3-8b-tmax-sft-v3-iter353",
		"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 "willamazon1/qwen3-8b-tmax-sft-v3-iter353" \
        --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": "willamazon1/qwen3-8b-tmax-sft-v3-iter353",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use willamazon1/qwen3-8b-tmax-sft-v3-iter353 with Docker Model Runner:
```
docker model run hf.co/willamazon1/qwen3-8b-tmax-sft-v3-iter353
```

Qwen3-8B TMax SFT v3 — iteration 353

Supervised fine-tuning (SFT) checkpoint of Qwen/Qwen3-8B, taken at training iteration 353 of the tmax_sft_v3 run.

This checkpoint is the initialization (policy + reference model) for the tmax_aenv_v39b reinforcement-learning run, whose checkpoints are published separately.


Base model	`Qwen/Qwen3-8B`
Stage	SFT
Iteration	353
Architecture	Qwen3, 36 layers, hidden 4096, 32 heads / 8 KV, vocab 151936
Precision	bfloat16

Usage

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

repo = "willamazon1/qwen3-8b-tmax-sft-v3-iter353"
tok = AutoTokenizer.from_pretrained(repo)
model = AutoModelForCausalLM.from_pretrained(repo, dtype=torch.bfloat16, device_map="auto")

msgs = [{"role": "user", "content": "What is 12*8?"}]
text = tok.apply_chat_template(msgs, tokenize=False, add_generation_prompt=True)
ids = tok(text, return_tensors="pt").input_ids.to(model.device)
out = model.generate(ids, max_new_tokens=256)
print(tok.decode(out[0][ids.shape[1]:], skip_special_tokens=True))

Conversion

Converted from a Megatron-LM torch_dist training checkpoint to HuggingFace safetensors with slime's tools/convert_torch_dist_to_hf.py. Embedding padding was stripped back to the tokenizer's vocab_size of 151936, so the tensor shapes match upstream Qwen/Qwen3-8B exactly. Weights are bfloat16; the optimizer state of the training checkpoint is not included.

Every shard was checked for NaN/Inf (none found) and the model was loaded and sampled from before upload.