OmniCoder-9B is a 9-billion parameter coding agent model built by
Tesslate, fine-tuned on top of
Qwen3.5-9B's hybrid architecture (Gated Delta Networks interleaved with standard attention). It was trained on
425,000+ curated agentic coding trajectories spanning real-world software engineering tasks, tool use, terminal operations, and multi-step reasoning.
The model shows strong agentic behavior: it recovers from errors (read-before-write), responds to LSP diagnostics, and uses proper edit diffs instead of full rewrites. These patterns were learned directly from the real-world agent trajectories it was trained on.
1from transformers import AutoModelForCausalLM, AutoTokenizer
2
3model_id = "Tesslate/OmniCoder-9B"
4
5tokenizer = AutoTokenizer.from_pretrained(model_id)
6model = AutoModelForCausalLM.from_pretrained(model_id, torch_dtype="auto", device_map="auto")
7
8messages = [
9 {"role": "system", "content": "You are a helpful coding assistant."},
10 {"role": "user", "content": "Write a Python function to find the longest common subsequence of two strings."},
11]
12
13text = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
14inputs = tokenizer([text], return_tensors="pt").to(model.device)
15
16outputs = model.generate(**inputs, max_new_tokens=2048, temperature=0.6, top_p=0.95, top_k=20)
17print(tokenizer.decode(outputs[0][inputs.input_ids.shape[-1]:], skip_special_tokens=True))
1from openai import OpenAI
2
3client = OpenAI(base_url="http://localhost:8000/v1", api_key="token")
4response = client.chat.completions.create(
5 model="Tesslate/OmniCoder-9B",
6 messages=[{"role": "user", "content": "Explain the difference between a mutex and a semaphore."}],
7 temperature=0.6,
8)
9print(response.choices[0].message.content)
For agentic / tool-calling tasks, consider lower temperature (0.2-0.4) for more deterministic behavior.
Special thanks to the
Axolotl team and the discussion in
axolotl#3453 for helping get Qwen3.5 packing support working.
1@misc{omnicoder2025,
2 title={OmniCoder-9B: A Frontier Open Coding Agent},
3 author={Tesslate},
4 year={2025},
5 url={https://huggingface.co/Tesslate/OmniCoder-9B}
6}
