Competitive programmers and researchers wanting a small, free coding model can run it themselves; general users won't need it.
Kenning's take
What it does
NousCoder-14B is Qwen3-14B further trained by Nous Research with reinforcement learning on 24,000 verifiable coding problems, aimed at competitive programming. Nous reports 67.87% Pass@1 on LiveCodeBench v6, up from 60.79% for the base model. The weights are free on Hugging Face under Apache 2.0.
- Post-trained from Qwen3-14B with reinforcement learning
- 67.87% Pass@1 on LiveCodeBench v6 (maker's figure)
- Trained on 24,000 verifiable coding problems
- Supports tool calling through its chat template
- Apache 2.0 open weights
Who it suits. Researchers and developers who want a small open model for algorithmic coding problems.
Key facts
- Category
- AI models
- Made by
- Nous Research
- Launched
- 2025
- How to get it
- free download from Hugging Face for self-hosting
- Works on
- self-hosted
- In the UK
- Yes
- Free plan
- free open weights
FAQ
What is NousCoder-14B?
An open 14B coding model from Nous Research, trained from Qwen3-14B for competitive programming.
How good is NousCoder-14B?
Nous Research reports 67.87% Pass@1 on LiveCodeBench v6, 7.08 points above Qwen3-14B.
Is NousCoder-14B free?
Yes. The weights are on Hugging Face under the Apache 2.0 licence.
Facts from the maker's own pages and the sources listed. Checked 2026-10-11.
Similar tools
ACE-Step 1.5
Open-source music generation model that writes full songs with lyrics and runs on ordinary consumer GPUs.
AlphaGenome Atlas
Google DeepMind database of AI-predicted effects for every possible single-letter change in human DNA.
Amazon Nova 2
Amazon's second family of Nova AI models, covering reasoning, speech conversation and combined text-and-image generation.
Cisco Antares
Cisco's small open-weight models that find where known security vulnerabilities sit in a codebase, running locally.