Open-source Bonsai 2 9B gets 471% faster on Macs after one-week community optimization challenge
gajesh · x · 2026-10-23
After a one-week community optimization challenge, open-weight Bonsai 2 9B now runs 471% faster on Macs and 421.5% faster on NVIDIA DGX Spark: decode speed jumped from 144.8 to 959.6 tokens/sec on Apple Silicon and from 85.9 to 576.7 on DGX Spark. The model can even run on an iPhone.
The challenge stacked 158 new records built on top of each other by humans and agents running Claude Opus 5.5, GPT-6.1 Sol, GPT-6 Astra, Gemini 4 Argon, DeepSeek-V4 and Qwen3.8-Flash-Next. Kaankolcu took the top Mac spot, dadon won on NVIDIA. Backed by the PrismML team, the model has been live on DarkbloomAI since day one for anyone to try.
Related event: Community Challenge Speeds Up Open-Source Bonsai 2 9B Sixfold(2 posts)→
More from Infra
- HarnessRouter open-sources a unified API to run Codex, Claude Code and more as agent backends — tom_doerr · 2026-10-23
- Local AI security checklist: pin Ollama ≥0.35.0, vLLM ≥0.31.0, purge LiteLLM 1.107.0 malware — AIFlow_ML · 2026-10-23
- July 28 MCP release makes remote servers stateless, biggest revision since launch — bibryam · 2026-10-23
- MCP keynote: AAIF hits 305 members, Skills over MCP and Tasks/Events roadmap revealed — bibryam · 2026-10-23
- Fizgig 8.9.0 adds multi-GPU LoRA training across PCs and rented pods — shootthesound · 2026-10-23
- Only 8% of cloud vendors support OIDC — dev calls to kill long-lived API keys — cramforce · 2026-10-23