Kimi K3.1 Pro benchmarks leak: tops SWE Marathon and 5 other agentic evals, near Opus 5.5
Ferzelibey · reddit · 2026-10-24
A Reddit post shares benchmark charts for Kimi K3.1 Pro, showing the open-source model leading or matching closed frontier models across many agentic evals:
- #1 results: SWE Marathon 51.7, Agents' Last Exam 43.2, OfficeQA Pro v2 64.4, Humanity's Last Exam Full w/ tools 79.0, GDP.pdf 34.6, ZeroBench w/ tools pass@5 63.0 — all above Opus 5.5 and GPT-6 Astra
- Near the top: Terminal-Bench 4.0 at 65.7 (just behind Opus 5.5's 66.4); GDPval-AA v2.1 at 1846; BenchCAD w/ tools 92.7
- Weaker spots: Terminal-Bench Science 0.1 at 38.1 vs GPT-6 Astra's 68.1; OSWorld V2.1 at 79.6 vs Opus 5.5's 91.8
Note: all models ran at max thinking effort. Overall, open-source now genuinely closes the gap on agentic tasks.
Related event: Moonshot Releases Kimi K3.1 Pro, Open-Weight Model Rivaling GPT-6(33 posts)→
More from Models
- Kimi 3.1 Pro priced like Argon but scores like DeepSeek V4.1 Flash on meme 'Runescape Bench' — scaling01 · 2026-10-24
- Jon Barron details his workflow: Gemini runs experiments, drafts the paper, then rewrites in his style — jon_barron · 2026-10-24
- Users praise Gem Argon for cultural fluency: a frontier model that roots for you — sebkrier · 2026-10-24
- User says Opus silently downgraded to 4.8 mid-task, then told him to talk to people — natesiggard · 2026-10-24
- Suspicious new GitHub accounts reference Qwen4 in Alibaba T-Head repo PRs — brainExploded99 · 2026-10-24
- Kimi K3.1 Pro takes on the pelican-on-a-bicycle 3D test: beats Opus 5, trails 5.5 — blaind0 · 2026-10-24