OSS Radar
nace-ai/kapitsa
Blazing fast Apple Silicon inference engine optimized for Qwen & Nemotron. OpenAI + Anthropic API compatible. Built by folks from nace.ai. Why it matters: useful plumbing for shipping LLM-powered features.
GitHub description: Blazing fast Apple Silicon inference engine optimized for Qwen & Nemotron. OpenAI + Anthropic API compatible. Built by folks from nace.ai
Stars
698
Last 24h
+30est.
Last 7 days
+208est.
Forks
102
Related repos
All kapitsa alternatives →- omlxjundot/omlx+184 stars this week
LLM inference server with continuous batching & SSD caching for Apple Silicon — managed from the macOS menu bar.
- HyperQwensyv-ai/HyperQwen+138 stars this week
Serve large Qwen models fast on the GPUs you actually own. Qwen3.8-27B on a single 24 GB card with vLLM…
- PolyStratasimplexai-labs/PolyStrata+155est. stars this week
Run large AI models on your own PC: Qwen3.8-Flash-Next on Strata's engine, and GLM 5.3 Flash, DeepSeek V4 Flash, LFM2.5, Qwen3.6, Ornith…
- RustingStrataRustingStudio/RustingStrata+139est. stars this week
Pure-Rust MoE inference engine for Qwen3.5/3.6/3.8 GGUF. CPU+GPU hybrid experts, NVRTC CUDA kernels, OpenAI-compatible server.
Who it's for
- 351 tok/s on Qwen3.6 35B-A3B (M5 Max), the fastest decode in 33 of 38 model and Mac pairs.
- A 180B model on a 48 GB Mac, at 61 tok/s (M5 Pro). Its routed experts stream from the SSD, as many as the budget holds.
- A 16K follow-up after a restart starts in 0.16 s instead of 17 s. Conversations stay on the SSD (`--disk-cache`).
Star history
GitHub stars over time. A steeper line means the repo is gaining attention faster.
Last 3 hours · +4 stars
* / est. Estimated: until we have enough daily snapshots for a repo, its gain is total stars ÷ days since creation. Measured numbers replace estimates automatically.
Discussed on Hacker News
Hacker News is a popular tech discussion site. 1 story links to this repo, with 1 upvotes (points) in total.
Apple Silicon Inference Engine: Kapitsa1 points · 0 comments · Oct 24, 2026Maintainer? Add the badge to your README
Shows your live OSS Radar rank and links back here. Updates hourly.
[](https://ossradar.net/repo/nace-ai/kapitsa)<a href="https://ossradar.net/repo/nace-ai/kapitsa"><img src="https://ossradar.net/badge/nace-ai/kapitsa" alt="Trending on OSS Radar"></a>About kapitsa
nace-ai/kapitsa is a C++ project in the LLM tooling category, created 23 days ago. It has 698 GitHub stars and 102 forks, and gained +208 stars over the last 7 days (early estimate), ranking #115 of 149 LLM tooling repos tracked by OSS Radar this week.
See more trending LLM tooling repos or the top AI repos this week. Looking for something similar? See kapitsa alternatives.
- Language
- C++
- Category
- LLM tooling
- License
- Apache-2.0
- Created
- Oct 16, 2026
- Last code push
- today (Nov 8, 2026)
- Latest release
- v0.2.31 2 days ago
- apple silicon
- clover
- coding agents
- decision models
- drex
- drex max
- fast local inference
- flint
- llama.cpp
- LLM inference
- local LLM
- macOS
- metal
- MLX
- nemotron
- Ollama
- OpenAI API
- Qwen
- slm
- speculative decoding
Rising in LLM tooling this week
- unslopheydamir/unslop+6.8k stars this week
A Claude skill that turns vibecoded, AI-looking UI into a designed interface.
- Claude-Memthedotmack/claude-mem+5.5k stars this week
Persistent Context Across Sessions for Every Agent – Captures everything your agent does during sessions, compresses it with AI, and injects…
- StrataNiko1221/Strata+4.1k stars this week
Qwen3.8-Flash-Next on any consumer hardware: one-click install for Windows / Linux.
- MoneyPrinterTurboharry0703/MoneyPrinterTurbo+3.7k stars this week
Generate HD short videos from a topic or keyword with an automated AI workflow.
- GraphifyGraphify-Labs/graphify+3.7k stars this week
Turn any codebase, with its docs, SQL schemas, configs, and PDFs, into a queryable knowledge graph.
- lemo-opuscarlemomo-ai/lemo-opuscar+3k stars this week
A toolbox of film styles made in code for Claude Code: 66 ready to use, and your own.
kapitsa is #277 of 450 on today's leaderboard.