Leaked Opus 5 Benchmarks: Major Jumps in Research Math and Long-Context
echen · x · 2026-07-30
Alleged benchmarks for the unreleased Opus 5 model have surfaced. Compared to Opus 4.8, the new version shows massive leaps in research math (Riemann-bench: 68.0% vs 47.2%), chart understanding (Chartography: 27.3% vs 15.9%), and long-context agents (HANDBOOK.md: 32.3% vs 21.9%). It shows modest improvements or ties in enterprise complex instructions, document reasoning, and creative writing.
Related event: Leaked Benchmarks Show Opus 5 Performance Leap(2 posts)→
More from Models
- Reflection's Beam, a 501B-parameter open MoE model with 1M context, lands on Hugging Face under Apache 2.0 — QGallouedec · 2026-10-16
- Why are sub-15B open models so rare now? Labs seem to have settled on 24-27B minimum — Hot_Example_4456 · 2026-10-16
- User spots model openly generating "ideal responses" to train other models — liminal_bardo · 2026-10-16
- Reflection's 501B MoE Beam hits 463 tok/s on 8x GB200 with NVFP4 at day 0 — zhyncs42 · 2026-10-16
- Gemma 4 31B QAT on one 24GB card: 46 tok/s plain, ~80 with draft model on RTX 4090 — Subject_Poet_3861 · 2026-10-16
- JevBench v1.7.46 leaderboard: new d1-ultra tops 32 API systems at 76.1 composite — airesearch12 · 2026-10-16