Instructions to use Montalte/qwen4b-code-think-taskvector with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Montalte/qwen4b-code-think-taskvector with Transformers:

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

pipe = pipeline("text-generation", model="Montalte/qwen4b-code-think-taskvector")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Montalte/qwen4b-code-think-taskvector")
model = AutoModelForCausalLM.from_pretrained("Montalte/qwen4b-code-think-taskvector", 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=256)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Montalte/qwen4b-code-think-taskvector with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Montalte/qwen4b-code-think-taskvector"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Montalte/qwen4b-code-think-taskvector",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Montalte/qwen4b-code-think-taskvector

SGLang

How to use Montalte/qwen4b-code-think-taskvector 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 "Montalte/qwen4b-code-think-taskvector" \
    --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": "Montalte/qwen4b-code-think-taskvector",
		"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 "Montalte/qwen4b-code-think-taskvector" \
        --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": "Montalte/qwen4b-code-think-taskvector",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Montalte/qwen4b-code-think-taskvector with Docker Model Runner:
```
docker model run hf.co/Montalte/qwen4b-code-think-taskvector
```

qwen4b-code-think-taskvector / provenance /DEV256_COMPLETE.json

Montalte

Upload qwen4b taskvector merge

f82d492 verified about 2 months ago

Raw History Blame Contribute Delete

5.92 kB

{"adapter":{"adapter_identity":"MERGED_FULL_WEIGHTS_NO_LORA","adapter_path":"/workspace/code-sft-runs/t30b2507-q4b-tn-v2-lr1e4-e1-20260907/arms/Q4B-THINK-MIX/out/Q4B-THINK-MIX/906bfd4b4dc7f14ee4320094d8b41684abff8539/qwen3-8b-think-lcb-system-v2/fc22e609554e898f2ae570f75bce208ae3428aada9931dc43bc7da3735f95885/step-000920-tokens-64848450/adapter","adapter_tree_sha256":"467d7c210b6001b8a1c638cce88ebb5112fa2f7d1684000f712b4ccafa814b29","checkpoint":"/workspace/code-sft-runs/t30b2507-q4b-tn-v2-lr1e4-e1-20260907/arms/Q4B-THINK-MIX/out/Q4B-THINK-MIX/906bfd4b4dc7f14ee4320094d8b41684abff8539/qwen3-8b-think-lcb-system-v2/fc22e609554e898f2ae570f75bce208ae3428aada9931dc43bc7da3735f95885/step-000920-tokens-64848450","checkpoint_manifest_sha256":"490367bf1e2bd55f5a7381dc4083bcffc18430bcbc26457680c5c17268225bcc","merged_path":"/workspace/code-sft-runs/rstar-eval/official-dev256/v4-q4b-think-mix/scratch/step-000920-tokens-64848450-merged-full"},"arm":"official-v4-q4b-think-mix","base":false,"benchmark_contract":{"date_filter":null,"native_index":{"bytes":2546647101,"immutable_object_uri":"r2mtx:mxx/isolated-backups/full-workspace/20260826T031752Z-8bcbe83efc6607702f6b681b4429d72dc6c1a4c570f18c85e8c7cee5743c682a/objects/sha256/ea/ea66e0d8814490a7e0d7ff555297e43a7d9662405c16609f9df51b42ec1b1cc1","path":"/workspace/code-sft-runs/rstar-eval/contracts/dev256/native_index.jsonl","sha256":"ea66e0d8814490a7e0d7ff555297e43a7d9662405c16609f9df51b42ec1b1cc1"},"ordered_task_ids_sha256":"8f4b14d4af19399ed59a052ef970ad18eb83f19cc5257bcafa4fafd1938315b9","prompts":{"path":"/workspace/code-sft-runs/rstar-eval/contracts/dev256/evalscope_prompts.jsonl","sha256":"b8c2047bf804c01a1cc82d16e07599dea0f4d94a42af6a26386c832a33747695"},"recovery":{"path":"/workspace/code-sft-runs/rstar-eval/contracts/dev256/RECOVERY.json","sha256":"011d8adba4986651a5d83add11626f155270008c9e69e89a4a86e753d71b0906"},"role":"DIAGNOSTIC_ONLY","rows":256,"schema":"CTX_LCDIV_DEV256_RECOVERY_V2","seed":3407,"seed_count":1,"subset_filter":null,"universe_rule":"exact content-addressed ordered task IDs recovered from six frozen test*.jsonl shards; no resampling, shuffle, date filtering, subset filtering, or hub refresh"},"cap_rate":0.53125,"caps":136,"checkpoint":"/workspace/code-sft-runs/t30b2507-q4b-tn-v2-lr1e4-e1-20260907/arms/Q4B-THINK-MIX/out/Q4B-THINK-MIX/906bfd4b4dc7f14ee4320094d8b41684abff8539/qwen3-8b-think-lcb-system-v2/fc22e609554e898f2ae570f75bce208ae3428aada9931dc43bc7da3735f95885/step-000920-tokens-64848450","code_plus_stop":73,"first_stop_accounted":256,"first_stop_token":{"100257":0,"100265":0,"151643":0,"151645":0,"length":136,"other":120},"gpu":2,"kind":"merged_full_weights","lora":false,"merged":"/workspace/code-sft-runs/rstar-eval/official-dev256/v4-q4b-think-mix/scratch/step-000920-tokens-64848450-merged-full","mode":"think","native_validation":{"modes":{"functional":109,"stdin":147},"native_index_sha256":"ea66e0d8814490a7e0d7ff555297e43a7d9662405c16609f9df51b42ec1b1cc1","ordered_evaluation_samples_sha256":"811cc4cd88b1863ab973be0b906c12934f1acc67d436f1b48d4a6863f7e0a4c7","ordered_task_ids_sha256":"8f4b14d4af19399ed59a052ef970ad18eb83f19cc5257bcafa4fafd1938315b9","rows":256},"pass_at_1":0.28515625,"passed":73,"pending_infra":0,"per_item":{"document_sha256":"ea2d8faff8649d26c33a83b40fea63e74221e4ce76f22c052259e721495b4b8b","path":"/workspace/code-sft-runs/rstar-eval/official-dev256/v4-q4b-think-mix/results/PER_ITEM.json","sha256":"c21c2587b33b0c6d2ed14da02c463fc26e57a3c64456f4e6c82c9ca8a45cc98a"},"port":18617,"protocol_failure":0,"queue_sha256":"80a75fca2a228428ecac539eb773958b8c38975d9adb727093548aefa27fa39b","receipt_sha256":"2e587612516917f00ff268d1309f1533cb6cb282dc1b25b4ed3170517df70036","renderer_authority":{"path":"/workspace/code-sft-runs/t30b2507-q4b-tn-v2-lr1e4-e1-20260907/tokenizers/qwen3-v4-tail151643.authority.json","sha256":"6b1461404d033b79f05a8cbb373f05a495d506a567b35b18cbf89930f345422a"},"renderer_id":"qwen3-8b-think-lcb-system-v2","rows":256,"runner":{"evalscope_adapter_sha256":"6c4977439b751003d6900c5cec2e7b3244599d3427903f99e3d7300d2c0d7687","evalscope_core_files":{"evaluate_utils.py":"a16167430f636906d15c6540f4640bb72973d9c34cea87928ebe870b51fc9794","extract_utils.py":"1c74c40a9250191cdaa5560104df583a4bec493154a2ea3eb603c35237c2a038","load_utils.py":"7c4cd6cff7d41b08b935cba6e4f35517d8c572e0f01a0b3b56dff5d182811364","testing_util.py":"72ac3f87e288cf39ed882e6f31cfc1192e5027d97d7d9a7730e2aa946effa329"},"evalscope_sandbox_sha256":"fcbbd6ec7634225ce023da194d178206ef778ad8e945e0ac36353934c09186f0","evalscope_version":"1.11.0","library_sha256":"0ab605fc8f758b1507752f19b4266c0a17e562774b1dcbe8e138708fd76ad7df","official_task":{"base":false,"cache_root":"/workspace/code-sft-runs/rstar-eval/official-dev256/v4-q4b-think-mix/cache","checkpoint":"/workspace/code-sft-runs/t30b2507-q4b-tn-v2-lr1e4-e1-20260907/arms/Q4B-THINK-MIX/out/Q4B-THINK-MIX/906bfd4b4dc7f14ee4320094d8b41684abff8539/qwen3-8b-think-lcb-system-v2/fc22e609554e898f2ae570f75bce208ae3428aada9931dc43bc7da3735f95885/step-000920-tokens-64848450","label":"v4-q4b-think-mix","mode":"think","student_family":"qwen"},"runner_id":"code-sft-rstar-lcb-handoff-v2","runner_sha256":"5ea2f032337a72c97f0c8cda2310e9f99e649f2bbe8cbcea8e2bcbd5072c6431","sandbox_runtime":{"execution_module_sha256":"cb162f70bb548694ad08372c71399b9011500b198cc7d23bf5dc2f95912189fc","pool_config_sha256":"33de3626e6591d5dcb0b1b99bb8fa53b05acd42a8a9fe65637e2fd2c890b0790","pool_probe_sha256":"6b2a91c4f5ff066ca9617602d166f4e19a9d5228a990ea2aa62b6fd76959ebc0","scope_lease":"CODE_SFT_SANDBOX_SCOPE_LEASE_V1","scope_module_sha256":"d5c0cb716af86aa2bc03c12cce8106046692672eafe6fb7cc7562e4c9752a8bf","transport":"SANDBOX_API_V1_DIRECT_TOOL_EXECUTE"}},"schema":"CODE_SFT_RSTAR_EVAL_SUITE_RECEIPT_V1","scored":256,"seed":3407,"served_model":"v4-q4b-think-mix-dev256","status":"COMPLETE","student_family":"qwen","suite":"dev256","unscored":0}