Qwen Pruning Tests: 256 Experts Optimal, Q4 Beats Q2
EyalToledano · x · 2026-08-27
Built a pruning curve for Qwen3.8-Flash-next to analyze size vs. accuracy tradeoffs:
- Sweet Spot: 256 experts is the optimal balance, halving size before performance buckles.
- Strategy Matters: Randomly cutting experts caused the model to collapse; selecting survivors is crucial.
- Quantization: Q2 quantization (full experts, 63GB) produced garbage output. Q4 quantization with half the experts performs significantly better.
More from Models
- DeepSeek outage compensation reportedly hands single accounts over $850K in API credits — teortaxesTex · 2026-10-15
- Nat Lambert: wave of AI releases rumored — thinky, muse spark, Fable 5.5, OpenAI — natolambert · 2026-10-15
- Musk amplifies dev's verdict: Grok Bot easier to use than Codex and Claude — elonmusk · 2026-10-15
- Rumor: New Anthropic model imminent, apparently behind a batch of high-severity Chrome CVEs — matthew_d_green · 2026-10-15
- StartLux-Decision debuts RL-trained calibrated decision models, 27B scores 72.8 — startlux-models · 2026-10-15
- Meta's Muse Code slammed for harsh quotas: 21% left after 5 hours — ThomasScialom · 2026-10-15