Moonshot ships Kimi K3.1 Pro: 3T params, near-Opus coding scores and a new BenchCAD benchmark
karminski3 · x · 2026-10-24
Moonshot AI released Kimi-K3.1-Pro with 3T parameters. Official benchmarks place its coding between Opus-5.5 and Fable-5.1, matching OpenAI/Anthropic flagships on agentic coding (SWE series) and setting a new high on TerminalBench. A new benchmark, BenchCAD (1,000-file subset, Voxe1 IoU), tests CAD modeling: without tools Kimi K3.1 Pro scores 72.8 vs Fable 5.1's 60.6; with tools it hits 92.7, near Opus 5.5 (96.2) and GPT-6 Astra (95.9). The launch blog shows elaborate CAD demos, including an RX-7 FD3S teardown with 1,153 individual parts. Pricing not yet clarified.
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