Instructions to use aasim-m/daft-qwen2.5-coder-7b-instruct-checkpoint-3500 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use aasim-m/daft-qwen2.5-coder-7b-instruct-checkpoint-3500 with Transformers:

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

pipe = pipeline("text-generation", model="aasim-m/daft-qwen2.5-coder-7b-instruct-checkpoint-3500")
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("aasim-m/daft-qwen2.5-coder-7b-instruct-checkpoint-3500")
model = AutoModelForCausalLM.from_pretrained("aasim-m/daft-qwen2.5-coder-7b-instruct-checkpoint-3500", 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 aasim-m/daft-qwen2.5-coder-7b-instruct-checkpoint-3500 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "aasim-m/daft-qwen2.5-coder-7b-instruct-checkpoint-3500"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "aasim-m/daft-qwen2.5-coder-7b-instruct-checkpoint-3500",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/aasim-m/daft-qwen2.5-coder-7b-instruct-checkpoint-3500

SGLang

How to use aasim-m/daft-qwen2.5-coder-7b-instruct-checkpoint-3500 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 "aasim-m/daft-qwen2.5-coder-7b-instruct-checkpoint-3500" \
    --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": "aasim-m/daft-qwen2.5-coder-7b-instruct-checkpoint-3500",
		"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 "aasim-m/daft-qwen2.5-coder-7b-instruct-checkpoint-3500" \
        --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": "aasim-m/daft-qwen2.5-coder-7b-instruct-checkpoint-3500",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use aasim-m/daft-qwen2.5-coder-7b-instruct-checkpoint-3500 with Docker Model Runner:
```
docker model run hf.co/aasim-m/daft-qwen2.5-coder-7b-instruct-checkpoint-3500
```

DAFT Qwen2.5-Coder-7B-Instruct — checkpoint 3500

Latest saved checkpoint from full supervised fine-tuning for NVIDIA GPU assembly to AMD GPU assembly translation. Training was stopped by the user after logged step 3,595. This is the latest saved checkpoint, not an export of the unsaved step-3,595 state.

Base model: Qwen/Qwen2.5-Coder-7B-Instruct, revision c03e6d358207e414f1eca0bb1891e29f1db0e242.
Optimizer step: 3,500 of 10,219 planned steps; one epoch, all 163,495 eligible training pairs.
Dataset: ahmedheakl/daft-sm89-rdna-functions, revision a833e9887bb5e3623c52dc0ce57480784fd77ce6.
Context: 32,768 total prompt and response tokens; full fine-tuning, BF16, two H200s, ZeRO-3, effective batch 16.
Leakage policy: benchmarks_exact_v2, excluding only exact whitespace-normalized source OR target benchmark matches from training.
Validation: target cross-entropy on 281 eligible CASS pairs and 337 eligible Rodinia pairs. These suites are used for checkpoint selection, not independent final testing. Functional correctness has not been established.
Export includes model weights and tokenizer. DeepSpeed optimizer/RNG state remains in the local training checkpoint and is not included in this model repository.
Training code: https://github.com/aasim-m/DAFT-experiment-setup
W&B: https://wandb.ai/daft/daft-asplos/runs/9fd9f1100493

Input format

Use the tokenizer's chat template with a user message containing:

Translate the following NVIDIA GPU assembly function into corresponding AMD GPU assembly. Preserve the function's behavior. Return only the translated assembly, without explanations or Markdown fences.
<your NVIDIA assembly function>

Validation losses at step 3500

eval_qwen25_cass_validation_loss: 0.07435634732246399
eval_qwen25_rodinia_validation_loss: 0.15731622278690338

Downloads last month: 14

Safetensors

Model size

333k params

Tensor type

BF16

Model tree for aasim-m/daft-qwen2.5-coder-7b-instruct-checkpoint-3500

Base model

Qwen/Qwen2.5-7B

Finetuned

Qwen/Qwen2.5-Coder-7B

Finetuned

Qwen/Qwen2.5-Coder-7B-Instruct

Finetuned

(496)

this model