Loading benchmark catalogue
Compare performance benchmarks, coding intelligence, reasoning evaluations, context limits, and API economics between Kimi K3.1 Pro (Moonshot AI) and Claude Sonnet 5.5 (Anthropic).
| Metric / Capability | Kimi K3.1 Pro | Claude Sonnet 5.5 |
|---|---|---|
| Developer / Organization | Moonshot AI | Anthropic |
| Release Date | 2026-10-23 | 2026-09-28 |
| Deployment / Weights | Proprietary Cloud API | Proprietary Cloud API |
| Context Window | 1.0M tokens (1,048,576 tokens) | 1M tokens (1,000,000 tokens) |
| Input Pricing (/1M Tokens) | $2 | $2 |
| Output Pricing (/1M Tokens) | $10 | $10 |
| Benchmark / Domain | Kimi K3.1 Pro | Claude Sonnet 5.5 | Advantage |
|---|---|---|---|
Artificial Analysis Intelligence Index Composite index aggregating 10+ evaluations of general reasoning and knowledge. | 55.1% | 56.0% | Claude Sonnet 5.5 (+2%) |
Artificial Analysis Coding Agent Index v2.0 Composite index of coding-agent systems (model + harness) across FrontierCode v1.1, SWE-Atlas-QnA, and Terminal-Bench v4. | — | 59.3% | — |
SWE-bench Pro Harder, contamination-resistant successor to SWE-bench spanning multiple languages. | — | 81.3% | — |
Terminal-Bench 4.0 Sandboxed command-line tasks requiring shell and file-system operation. | 65.7% | 70.6% | Claude Sonnet 5.5 (+7%) |
ARC-AGI-2 Novel visual grid puzzles requiring abstract rule induction. | — | 83.9% | — |
GPQA Diamond Graduate-level physics, chemistry, and biology questions written by domain experts. | 94.2% | — | — |
Humanity's Last Exam Expert-written questions across hundreds of disciplines designed to resist saturation. | 51.1% | 56.9% | Claude Sonnet 5.5 (+11%) |
FrontierMath v2 (Tiers 1-3) Unpublished, research-level mathematics problems across Tiers 1-3. | — | 88.8% | — |
Design Arena ELO Human-preference ELO across UI, 3D, data-visualization, and game-dev generations. | 1448 | 1413 | Kimi K3.1 Pro (+2%) |
Claude Sonnet 5.5 outperforms Kimi K3.1 Pro across 3 evaluated benchmarks compared to 1 for Kimi K3.1 Pro, excelling in Artificial Analysis Intelligence Index, Terminal-Bench 4.0, Humanity's Last Exam.
Kimi K3.1 Pro offers a larger context capacity of 1.0M tokens (1,048,576 tokens), compared to 1M tokens for Claude Sonnet 5.5.
Both models have comparable pricing structures or open weight deployment options.
Kimi K3.1 Pro is a proprietary cloud API. Claude Sonnet 5.5 is a proprietary cloud API.
Analyze 130+ models side-by-side using interactive radar charts, cost-efficiency quadrant plots, and live dataset filters on the full NeuBench index.
Open Interactive Comparator