Independent eval puts Prism-LM's Bonsai 2 QAT models at ~91% accuracy
airesearch12 · x · 2026-09-19
ByteShape independently benchmarked Prism-LM's new Bonsai 2 QAT (quantization-aware training) models on the Qwen3.8 series, reporting roughly 91% — a solid result.
The numbers come from ByteShape's own runs using Prism's runtime, not Prism's reported benchmarks, with the rest of the setup consistent with their earlier Qwen3.8 evaluations.
Related event: Bonsai 2 QAT quantized model scores ~91.5% in independent benchmarks(2 posts)→
More from Models
- Musk cites unsustainable Grok Bot costs as Hermes offers one-command agent migration — Teknium · 2026-11-04
- Hobbyist rewrites Qwen 0.6B weights with 39.8M tokens, no gradients, in 3h12m — Altruistic_Heat_9531 · 2026-11-04
- Same leak, reshared: ChatGPT's five family plan prices spotted again — heyshrutimishra · 2026-11-04
- KOL visits StepFun's Shanghai office, hints a powerful new model is coming — teortaxesTex · 2026-11-04
- Fable 5.5 delayed as Anthropic reportedly bets OpenAI's Astra 6.2 won't beat it — haider1 · 2026-11-04
- ByteDance open-sources Axion 1.0-35B agent foundation model with full reproducible recipe — pmttyji · 2026-11-04