HF2Ollama. Download AI models from HuggingFace and run them locally with Ollama in one command.

pypi.org/project/hf2ollama
by @wachawo

Vaya's read on this project

Problem, audience, market, and the verdict — sign in to see it.

Updates

May 2026
  • Release —0.0.4
  • Merge pull request #10 from wachawo/gguf-metadata
  • fix: widen repeat_last_n window and bump repeat_penalty to break para…
  • docs: promote grounding system prompt as primary escalation for creat…
  • fix: fall back to GGUF metadata for chat_template and num_ctx
  • docs: rename ollama-cheatsheet.md -> OLLAMA.md
  • docs: explain sampling defaults rationale and escalation paths
  • docs: replace creative-tune author list with neutral description
  • feat: always emit SYSTEM block in Modelfile with grounding default
  • docs: rename sampling.md -> SAMPLING.md
  • docs: add Ollama runtime cheat sheet (REPL commands, env vars, KV-cac…
  • feat: CLI flags for sampling/context overrides, mirroring env var names
  • feat: add top_k + min_p sampler params; recalibrate defaults to Qwen3…
  • Merge pull request #6 from wachawo/modelfile-sampling-defaults
  • fix: write num_ctx + sampling defaults into generated Modelfile
  • Merge pull request #2 from wachawo/fix-modelfile-chat-template
  • fix: detect chat template family and emit TEMPLATE + PARAMETER stop
  • Release —0.0.3
  • Bump version to 0.0.3
  • feat: add --remove to wipe a model's HF snapshot and hub cache
  • docs: refresh .env.example with all HF2OLLAMA_* overrides
  • feat: introduce `hf2ollama init` and reuse already-active venvs
  • Release —0.0.2
  • Release —0.0.1
  • Release —0.0.0