Claude Opus 4.8vsKimi K2.6
Across 9 shared benchmarks, Claude Opus 4.8 leads overall: Claude Opus 4.8 wins 9, Kimi K2.6 wins 0, with 0 ties and an average score difference of +9.13.
Claude Opus 4.8
Anthropic · 2026-05-28 · Reasoning model
Kimi K2.6
Moonshot AI · 2026-04-20 · Reasoning model
Claude Opus 4.89 wins(100%)(0%)0 winsKimi K2.6
Benchmark scores
Grouped by capability, sorted by largest gap within each. 9 shared benchmarks.
AI Agent - Tool Usage
Claude Opus 4.8 3/3| Benchmark | Claude Opus 4.8 | Kimi K2.6 | Diff |
|---|---|---|---|
| Terminal-Bench 2.1 | 78.9020 / 44Thinking High (With Tools) | 53.5643 / 44Thinking (No Tools) | +25.34 |
| MCP-Atlas | 82.206 / 38Deep Thinking (With Tools) | 69.4028 / 38Thinking (With Tools) | +12.80 |
| OSWorld-Verified | 83.404 / 26Extended (with tools) | 73.1015 / 26Thinking (With Tools) | +10.30 |
Coding and Software Engineer
Claude Opus 4.8 2/2| Benchmark | Claude Opus 4.8 | Kimi K2.6 | Diff |
|---|---|---|---|
| SWE-Bench Pro - Public | 69.204 / 57Extended (with tools) | 58.6015 / 57Thinking (With Tools) | +10.60 |
| SWE-bench Verified | 88.605 / 114Extended (with tools) | 80.2014 / 114Thinking (With Tools) | +8.40 |
General Knowledge
Claude Opus 4.8 2/2| Benchmark | Claude Opus 4.8 | Kimi K2.6 | Diff |
|---|---|---|---|
| LiveBench | 78.794 / 115Deep Thinking (No Tools) | 72.1728 / 115Thinking (No Tools) | +6.62 |
| HLE | 57.908 / 181Extended (with tools) | 5417 / 181Thinking (With Tools + Internet) | +3.90 |
AI Agent - Information Search
Claude Opus 4.8 1/1| Benchmark | Claude Opus 4.8 | Kimi K2.6 | Diff |
|---|---|---|---|
| BrowseComp | 84.309 / 54Thinking High (With Tools + Internet) | 83.2014 / 54Thinking (With Tools + Internet) | +1.10 |
General Evaluation
Claude Opus 4.8 1/1| Benchmark | Claude Opus 4.8 | Kimi K2.6 | Diff |
|---|---|---|---|
| GPQA Diamond | 93.6011 / 226Thinking High (No Tools) | 90.5035 / 226Thinking (No Tools) | +3.10 |
Specs
| Field | Claude Opus 4.8 | Kimi K2.6 |
|---|---|---|
| Publisher | Anthropic | Moonshot AI |
| Release date | 2026-05-28 | 2026-04-20 |
| Model type | Reasoning model | Reasoning model |
| Architecture | Dense | MoE |
| Parameters | Not available | 1T |
| Context length | 1M | 256K |
| Max output | 125K | Not available |
API pricing
Prices use DataLearner records when available; missing fields are not inferred.
| Item | Claude Opus 4.8 | Kimi K2.6 |
|---|---|---|
| Text input | $5 / 1M tokens | $0.95 / 1M tokens |
| Text output | $25 / 1M tokens | $4 / 1M tokens |
| Cache read | $0.5 / 1M tokens | $0.16 / 1M tokens |
| Cache write | $6.25 / 1M tokens | $0.95 / 1M tokens |
Summary
- Claude Opus 4.8leads in:AI Agent - Tool Usage (3/3), Coding and Software Engineer (2/2), General Knowledge (2/2), AI Agent - Information Search (1/1), General Evaluation (1/1)
On average across the 9 shared benchmarks, Claude Opus 4.8 scores 9.13 higher.
Largest single-benchmark gap: Terminal-Bench 2.1 — Claude Opus 4.8 78.90 vs Kimi K2.6 53.56 (+25.34).
Page generated from structured model, pricing and benchmark records. No real-time LLM is used to write the prose.