deepcoder:14b

Upstream reference: DeepCoder 14B Preview

Publisher: Agentica and Together AI

The publisher describes this code-reasoning model as reinforcement-learning post-training of the corresponding DeepSeek-R1-Distill-Qwen checkpoint.

Publisher-described context, separate from Sigma5C observations and recorded catalog statistics.

Published upstream specifications

Published size label

14B Source

Upstream derivative license label

MIT Source

Base lineage

DeepSeek-R1-Distill-Qwen-14B Source

Publisher-described uses

• Code reasoning and generation under the documented research setup. Source

Reference limitations

• The preview is a derivative checkpoint, not an unmodified DeepSeek or Qwen release; upstream lineage and derivative terms should remain visible. Source

A recorded tag and an upstream card association do not verify the deployed checkpoint, quantization, configured context or current hosted availability.

Publisher sources

• DeepCoder 14B Preview official documentation — Agentica and Together AI. Checked 2026-11-02.

Links

Recorded references do not establish downloadable weights, licensing or current availability.

• Agentica / Together AI — official developer/publisher
• Ollama
• Hugging Face
• Official model announcement / documentation
• GitHub (maker)

Where deepcoder:14b comes from

Named in 2 observation pages

• 14B-32B local models — all 47 · V1
• Thinking-token re-observation — 14B-32B · V2