Model comparison
Claude Opus 4.5 vs Kimi K2.6
Head-to-head evidence from 36 shared benchmark results across 7 categories. Overall scores shown here use BenchLM's provisional ranking lane.
Verified leaderboard positions: Claude Opus 4.5 #19; Kimi K2.6 #13
Evidence parity. Claude Opus 4.5 and Kimi K2.6 share 36 comparable benchmark results. 5 of 8 categories are comparable. 24 results are unique to Claude Opus 4.5; 24 to Kimi K2.6.
Updated July 13, 2026- Shared results
- 36
- Claude Opus 4.5 only
- 24
- Kimi K2.6 only
- 24
- Comparable categories
- 5 / 8
Pick Kimi K2.6 if you want the stronger benchmark profile. Claude Opus 4.5 only becomes the better choice if knowledge is the priority or you would rather avoid the extra latency and token burn of a reasoning model.
Confidence note. This is a partial-evidence comparison with 36 shared benchmark results across 7 evidence categories; 5 of 8 categories currently have scoreable aggregates for both models. Treat the verdict as directional until coverage is more balanced.
Why this result
Kimi K2.6 has the cleaner provisional overall profile here, landing at 74 versus 72. It is a real lead, but still close enough that category-level strengths matter more than the headline number.
Kimi K2.6's sharpest advantage is in agentic, where it averages 73.5 against 62.6. The single biggest benchmark swing on the page is FrontierMath v2 (Tiers 1-3), 20.690% to 38.966%. Claude Opus 4.5 does hit back in knowledge, so the answer changes if that is the part of the workload you care about most.
Claude Opus 4.5 is also the more expensive model on tokens at $5.00 input / $25.00 output per 1M tokens, versus $0.95 input / $4.00 output per 1M tokens for Kimi K2.6. That is roughly 6.3x on output cost alone. Kimi K2.6 is the reasoning model in the pair, while Claude Opus 4.5 is not. That usually helps on harder chain-of-thought-heavy tests, but it can also mean more latency and more token spend in real use. Kimi K2.6 gives you the larger context window at 256K, compared with 200K for Claude Opus 4.5.
Category breakdown
Exact category averages are shown below. Not measured means BenchLM does not have enough sourced public coverage for that model and category.
| Category | Claude Opus 4.5 | Δ | Kimi K2.6 |
|---|---|---|---|
| Knowledge | Claude Opus 4.558.5 | Margin← 16.3 | Kimi K2.642.2 |
| Agentic | Claude Opus 4.562.6 | Margin→ 10.9 | Kimi K2.673.5 |
| Multimodal | Claude Opus 4.569.9 | Margin→ 9.9 | Kimi K2.679.8 |
| Math | Claude Opus 4.557.5 | Margin→ 9.6 | Kimi K2.667.1 |
| Coding | Claude Opus 4.566.0 | Margin→ 6.6 | Kimi K2.672.6 |
| Reasoning | Claude Opus 4.564.4 | MarginNo overlap | Kimi K2.6Not measured |
| Multilingual | Claude Opus 4.585.7 | MarginNo overlap | Kimi K2.6Not measured |
| Inst. Following | Claude Opus 4.569.5 | MarginNo overlap | Kimi K2.6Not measured |
Decisive benchmark drivers
The largest measured benchmark gaps in this matchup, with exact reported values.
More
- Source ↗
FrontierMath v2 (Tiers 1-3)
MathA 20.690%B 38.966%Winner: Kimi K2.6Δ 18.3FrontierMath v2 (Tiers 1-3): Claude Opus 4.5 scored 20.690%; Kimi K2.6 scored 38.966%. Kimi K2.6 wins this benchmark. - Source ↗
CharXiv
MultimodalA 68.5%B 80.4%Winner: Kimi K2.6Δ 11.9CharXiv: Claude Opus 4.5 scored 68.5%; Kimi K2.6 scored 80.4%. Kimi K2.6 wins this benchmark. - Source ↗
FrontierMath v2 (Tier 4)
MathA 4.167%B 14.580%Winner: Kimi K2.6Δ 10.4FrontierMath v2 (Tier 4): Claude Opus 4.5 scored 4.167%; Kimi K2.6 scored 14.580%. Kimi K2.6 wins this benchmark. - Source ↗
MMMU-Pro
MultimodalA 70.6%B 79.4%Winner: Kimi K2.6Δ 8.8MMMU-Pro: Claude Opus 4.5 scored 70.6%; Kimi K2.6 scored 79.4%. Kimi K2.6 wins this benchmark. - Source ↗
Terminal-Bench 2.0
AgenticA 59.3%B 66.7%Winner: Kimi K2.6Δ 7.4Terminal-Bench 2.0: Claude Opus 4.5 scored 59.3%; Kimi K2.6 scored 66.7%. Kimi K2.6 wins this benchmark.
Operational comparison
Runtime and commercial metrics are compared only when both models have a complete sourced value.
| Metric | Claude Opus 4.5 | Kimi K2.6 | Comparison |
|---|---|---|---|
| Input / output priceUSD per 1M tokens | Claude Opus 4.5$5 input / $25 output | Kimi K2.6$0.95 input / $4 output | Kimi K2.6 has the lower combined listed price. |
| Generation speedtokens per second | Claude Opus 4.546 tok/s | Kimi K2.6Not available | A complete speed comparison is not available. |
| First-answer latencyseconds to first token | Claude Opus 4.51.01 s | Kimi K2.6Not available | A complete latency comparison is not available. |
| Context windowmaximum listed tokens | Claude Opus 4.5200K | Kimi K2.6256K | Kimi K2.6 lists the larger context window. |
Benchmark Deep Dive
AgenticKimi K2.6 wins30 benchmarks
| Benchmark | Claude Opus 4.5 | Kimi K2.6 | Result |
|---|---|---|---|
| Terminal-Bench 2.0Source | 59.3% | 66.7% | Kimi K2.6 leads |
| OSWorld-VerifiedSource | 66.3% | 73.1% | Kimi K2.6 leads |
| OSWorldSource | 66.3% | — | Not comparable |
| Claw-EvalSource | 59.6% | 62.3% | Kimi K2.6 leads |
| QwenClawBenchSource | 52.3% | — | Not comparable |
| TAU3-BenchSource | 70.2% | — | Not comparable |
| VITA-BenchSource | 23.3% | — | Not comparable |
| DeepPlanningSource | 26.4% | — | Not comparable |
| ToolathlonSource | 43.5% | 50% | Kimi K2.6 leads |
| MCP AtlasSource | 42.3% | 55.9% | Kimi K2.6 leads |
| MCP-TasksSource | 71.8% | — | Not comparable |
| WideResearchSource | 76.4% | 80.8% | Kimi K2.6 leads |
| CyberGymSource | 50.6% | — | Not comparable |
| Tau2-TelecomSource | 86.3% | 95.9% | Kimi K2.6 leads |
| Gert LabsSource | 64.23% | 56.82% | Claude Opus 4.5 leads |
| JobBenchSource | 32.3% | — | Not comparable |
| BrowseCompSource | — | 83.2% | Not comparable |
| DeepSearchQASource | — | 92.5% | Not comparable |
| AA Agentic IndexSource | — | 30.3% | Not comparable |
| GDPval-AASource | — | 34.5% | Not comparable |
| GDPval-AASource | — | 1190 | Not comparable |
| APEX-Agents-AASource | — | 28.5% | Not comparable |
| ResearchClawBenchSource | — | 18.0% | Not comparable |
| OSWorld 2.0Source | — | 4.6% | Not comparable |
| AA BriefcaseSource | — | 809 | Not comparable |
| AA AutomationBenchSource | — | 19.6% | Not comparable |
| AA EnterpriseOps-GymSource | — | 38.5% | Not comparable |
| AA Harvey LABSource | — | 0.0% | Not comparable |
| AA ITBenchSource | — | 31.2% | Not comparable |
| AA Tau3 BankingSource | — | 20.6% | Not comparable |
CodingKimi K2.6 wins14 benchmarks
| Benchmark | Claude Opus 4.5 | Kimi K2.6 | Result |
|---|---|---|---|
| SWE-bench VerifiedSource | 80.9% | 80.2% | Claude Opus 4.5 leads |
| LiveCodeBench v6Source | 84.8% | 89.6% | Kimi K2.6 leads |
| SWE-bench ProSource | 57.1% | 58.6% | Kimi K2.6 leads |
| SWE MultilingualSource | 77.5% | 76.7% | Claude Opus 4.5 leads |
| NL2RepoSource | 43.2% | — | Not comparable |
| Terminal-Bench HardSource | 40.9% | 43.9% | Kimi K2.6 leads |
| AA-SciCodeSource | 47.0% | 53.5% | Kimi K2.6 leads |
| LiveCodeBenchSource | — | 89.6% | Not comparable |
| SciCodeSource | — | 52.2% | Not comparable |
| Terminal-Bench 2.0Source | — | 66.7% | Not comparable |
| Vibe Code BenchSource | — | 37.89% | Not comparable |
| cursorBench31Source | — | 47.6% | Not comparable |
| AA Coding IndexSource | — | 61.8% | Not comparable |
| AA Terminal-Bench 2.1Source | — | 65.9% | Not comparable |
Reasoning4 benchmarks
KnowledgeClaude Opus 4.5 wins15 benchmarks
| Benchmark | Claude Opus 4.5 | Kimi K2.6 | Result |
|---|---|---|---|
| GPQASource | 87% | 90.5% | Kimi K2.6 leads |
| SuperGPQASource | 70.6% | — | Not comparable |
| MMLU-ProSource | 89.5% | — | Not comparable |
| MMLU-ReduxSource | 96.6% | — | Not comparable |
| C-EvalSource | 92.2% | — | Not comparable |
| HLESource | 30.8% | 34.7% | Kimi K2.6 leads |
| Artificial Analysis Intelligence IndexSource | 34.7% | 44.2% | Kimi K2.6 leads |
| AA-GPQA DiamondSource | 81.0% | 91.1% | Kimi K2.6 leads |
| AA-HLESource | 12.9% | 35.9% | Kimi K2.6 leads |
| AA-Omniscience IndexSource | -3.9% | 6.4% | Kimi K2.6 leads |
| AA-Omniscience AccuracySource | 40.7% | 32.8% | Claude Opus 4.5 leads |
| AA-Omniscience Hallucination RateSource | 75.4% | 39.3% | Kimi K2.6 leads |
| AA MMLU-ProSource | 88.9% | — | Not comparable |
| GPQA-DSource | — | 90.5% | Not comparable |
| AA Openness IndexSource | — | 33.3% | Not comparable |
MathKimi K2.6 wins7 benchmarks
| Benchmark | Claude Opus 4.5 | Kimi K2.6 | Result |
|---|---|---|---|
| AIME26Source | 95.1% | 96.4% | Kimi K2.6 leads |
| HMMT Feb 2025Source | 92.9% | — | Not comparable |
| HMMT Nov 2025Source | 93.3% | — | Not comparable |
| HMMT Feb 2026Source | 85.3% | 92.7% | Kimi K2.6 leads |
| MMAnswerBenchSource | 84.0% | 86.0% | Kimi K2.6 leads |
| FrontierMath v2 (Tiers 1-3)Source | 20.690% | 38.966% | Kimi K2.6 leads |
| FrontierMath v2 (Tier 4)Source | 4.167% | 14.580% | Kimi K2.6 leads |
Multilingual2 benchmarks
MultimodalKimi K2.6 wins9 benchmarks
| Benchmark | Claude Opus 4.5 | Kimi K2.6 | Result |
|---|---|---|---|
| MMMU-ProSource | 70.6% | 79.4% | Kimi K2.6 leads |
| MathVisionSource | 74.3% | 87.4% | Kimi K2.6 leads |
| CharXivSource | 68.5% | 80.4% | Kimi K2.6 leads |
| VideoMMMUSource | 84.4% | — | Not comparable |
| ScreenSpot ProSource | 45.7% | — | Not comparable |
| V*Source | 67.0% | 96.9% | Kimi K2.6 leads |
| AA-MMMU-ProSource | 71.2% | 79.4% | Kimi K2.6 leads |
| Design Arena WebsiteSource | 1290 | 1318 | Kimi K2.6 leads |
| MMMU-Pro w/ PythonSource | — | 80.1% | Not comparable |
Frequently Asked Questions (6)
Which is better, Claude Opus 4.5 or Kimi K2.6?
Kimi K2.6 is ahead on BenchLM's provisional leaderboard, 74 to 72. The biggest single separator in this matchup is FrontierMath v2 (Tiers 1-3), where the scores are 20.690% and 38.966%.
Which is better for knowledge tasks, Claude Opus 4.5 or Kimi K2.6?
Claude Opus 4.5 has the edge for knowledge tasks in this comparison, averaging 58.5 versus 42.2. Inside this category, AA-Omniscience Hallucination Rate is the benchmark that creates the most daylight between them.
Which is better for coding, Claude Opus 4.5 or Kimi K2.6?
Kimi K2.6 has the edge for coding in this comparison, averaging 72.6 versus 66. Inside this category, AA-SciCode is the benchmark that creates the most daylight between them.
Which is better for math, Claude Opus 4.5 or Kimi K2.6?
Kimi K2.6 has the edge for math in this comparison, averaging 67.1 versus 57.5. Inside this category, FrontierMath v2 (Tiers 1-3) is the benchmark that creates the most daylight between them.
Which is better for agentic tasks, Claude Opus 4.5 or Kimi K2.6?
Kimi K2.6 has the edge for agentic tasks in this comparison, averaging 73.5 versus 62.6. Inside this category, MCP Atlas is the benchmark that creates the most daylight between them.
Which is better for multimodal and grounded tasks, Claude Opus 4.5 or Kimi K2.6?
Kimi K2.6 has the edge for multimodal and grounded tasks in this comparison, averaging 79.8 versus 69.9. Inside this category, V* is the benchmark that creates the most daylight between them.
Self-host vs API cost
Estimates at 50,000 req/day · 1000 tokens/req average.
Related Comparisons
The AI models change fast. We track them for you.
A weekly brief for engineers and researchers covering new models, ranking shifts, and pricing changes.
Free. No spam. Unsubscribe anytime.