Model comparison
Kimi K2.5 vs o3
Head-to-head evidence from 15 shared benchmark results across 7 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.
Public leaderboard positions: Kimi K2.5 #54 (Supported); o3 #131 (Supported). Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.
Evidence parity. Kimi K2.5 and o3 share 15 comparable benchmark results. 1 of 8 categories are comparable. 48 results are unique to Kimi K2.5; 1 to o3.
Updated July 20, 2026- Shared results
- 15
- Kimi K2.5 only
- 48
- o3 only
- 1
- Comparable categories
- 1 / 8
Pick Kimi K2.5 if you want the stronger benchmark profile. o3 only becomes the better choice if you want the stronger reasoning-first profile.
Confidence note. This is a partial-evidence comparison with 15 shared benchmark results across 7 evidence categories; 1 of 8 categories currently have scoreable aggregates for both models. Treat the verdict as directional until coverage is more balanced.
Why this result
Kimi K2.5 is clearly ahead on the BenchAlign aggregate, 59.66 to 47.89. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.
Kimi K2.5's sharpest advantage is in mathematics, where it averages 60.6 against 14.5. The single biggest benchmark swing on the page is FrontierMath v2 (Tiers 1-3), 27.900% to 18.685%.
o3 is also the more expensive model on tokens at $2.00 input / $8.00 output per 1M tokens, versus $0.60 input / $3.00 output per 1M tokens for Kimi K2.5. That is roughly 2.7x on output cost alone. o3 is the reasoning model in the pair, while Kimi K2.5 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 K2.5 gives you the larger context window at 256K, compared with 200K for o3.
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 | Kimi K2.5 | Δ | o3 |
|---|---|---|---|
| Math | Kimi K2.560.6 | Margin← 46.1 | o314.5 |
| Agentic | Kimi K2.555.0 | MarginNo overlap | o3Not measured |
| Coding | Kimi K2.559.4 | MarginNo overlap | o3Not measured |
| Reasoning | Kimi K2.561.0 | MarginNo overlap | o3Not measured |
| Knowledge | Kimi K2.556.9 | MarginNo overlap | o3Not measured |
| Multilingual | Kimi K2.582.3 | MarginNo overlap | o3Not measured |
| Multimodal | Kimi K2.578.5 | MarginNo overlap | o3Not measured |
| Inst. Following | Kimi K2.593.9 | MarginNo overlap | o3Not measured |
Decisive benchmark drivers
The largest measured benchmark gaps in this matchup, with exact reported values.
More
- Source ↗
FrontierMath v2 (Tiers 1-3)
MathA 27.900%B 18.685%Winner: Kimi K2.5Δ 9.2FrontierMath v2 (Tiers 1-3): Kimi K2.5 scored 27.900%; o3 scored 18.685%. Kimi K2.5 wins this benchmark. - Source ↗
FrontierMath v2 (Tier 4)
MathA 4.200%B 2.083%Winner: Kimi K2.5Δ 2.1FrontierMath v2 (Tier 4): Kimi K2.5 scored 4.200%; o3 scored 2.083%. Kimi K2.5 wins this benchmark.
Operational comparison
Runtime and commercial metrics are compared only when both models have a complete sourced value.
| Metric | Kimi K2.5 | o3 | Comparison |
|---|---|---|---|
| Input / output priceUSD per 1M tokens | Kimi K2.5$0.6 input / $3 output | o3$2 input / $8 output | Kimi K2.5 has the lower combined listed price. |
| Generation speedtokens per second | Kimi K2.545 tok/s | o3118 tok/s | o3 has the higher measured throughput. |
| First-answer latencyseconds to first token | Kimi K2.52.38 s | o35.38 s | Kimi K2.5 reaches the first token sooner. |
| Context windowmaximum listed tokens | Kimi K2.5256K | o3200K | Kimi K2.5 lists the larger context window. |
Benchmark Deep Dive
Agentic19 benchmarks
| Benchmark | Kimi K2.5 | o3 | Result |
|---|---|---|---|
| Terminal-Bench 2.0Source | 50.8% | — | Not comparable |
| BrowseCompSource | 60.6% | — | Not comparable |
| Claw-EvalSource | 52.3% | — | Not comparable |
| QwenClawBenchSource | 54.3% | — | Not comparable |
| τ³-bench resultsSource | 65.7% | — | Not comparable |
| DeepSearchQASource | 77.1% | — | Not comparable |
| DeepPlanningSource | 14.4% | — | Not comparable |
| ToolathlonSource | 27.8% | — | Not comparable |
| MCP AtlasSource | 29.5% | — | Not comparable |
| MCP-TasksSource | 59.1% | — | Not comparable |
| WideResearchSource | 72.7% | — | Not comparable |
| τ²-bench resultsSource | 95.9% | 80.7% | Kimi K2.5 leads |
| APEX-Agents-AASource | 11.5% | — | Not comparable |
| Gert LabsSource | 45.88% | — | Not comparable |
| ResearchClawBenchSource | 14.0% | — | Not comparable |
| JobBenchSource | 8.7% | — | Not comparable |
| AA Agentic IndexSource | 21.7% | — | Not comparable |
| GDPval-AASource | 25.4% | — | Not comparable |
| GDPval-AASource | 1009 | — | Not comparable |
Coding10 benchmarks
| Benchmark | Kimi K2.5 | o3 | Result |
|---|---|---|---|
| SWE-bench VerifiedSource | 76.8% | — | Not comparable |
| SWE-bench Verified*Source | 70.8% | — | Not comparable |
| LiveCodeBench v6Source | 85.0% | — | Not comparable |
| SWE-bench ProSource | 50.7% | — | Not comparable |
| SWE MultilingualSource | 73% | — | Not comparable |
| SWE-RebenchSource | 58.5% | — | Not comparable |
| React Native EvalsSource | 77.2% | — | Not comparable |
| SciCodeSource | 48.7% | — | Not comparable |
| AA-SciCodeSource | 49.0% | 41.0% | Kimi K2.5 leads |
| AA Coding IndexSource | 46.8% | — | Not comparable |
Reasoning3 benchmarks
Knowledge12 benchmarks
| Benchmark | Kimi K2.5 | o3 | Result |
|---|---|---|---|
| GPQASource | 87.6% | — | Not comparable |
| GPQA-DSource | 87.6% | — | Not comparable |
| SuperGPQASource | 69.2% | — | Not comparable |
| MMLU-ProSource | 87.1% | — | Not comparable |
| MMLU-Pro (Arcee)Source | 87.1% | — | Not comparable |
| HLESource | 30.1% | — | Not comparable |
| Artificial Analysis Intelligence IndexSource | 35.4% | 30.4% | Kimi K2.5 leads |
| AA-GPQA DiamondSource | 87.9% | 82.7% | Kimi K2.5 leads |
| AA-HLESource | 29.4% | 20.0% | Kimi K2.5 leads |
| AA-Omniscience IndexSource | -8.1% | -15.3% | Kimi K2.5 leads |
| AA-Omniscience AccuracySource | 34.3% | 38.4% | o3 leads |
| AA-Omniscience Hallucination RateSource | 64.6% | 87.1% | Kimi K2.5 leads |
MathKimi K2.5 wins10 benchmarks
| Benchmark | Kimi K2.5 | o3 | Result |
|---|---|---|---|
| AIME 2025Source | 96.1% | — | Not comparable |
| AIME26Source | 95.8% | — | Not comparable |
| AIME25 (Arcee)Source | 96.3% | — | Not comparable |
| HMMT Feb 2025Source | 95.4% | — | Not comparable |
| HMMT Nov 2025Source | 91.1% | — | Not comparable |
| HMMT Feb 2026Source | 87.1% | — | Not comparable |
| MMAnswerBenchSource | 81.8% | — | Not comparable |
| FrontierMath v2 (Tiers 1-3)Source | 27.900% | 18.685% | Kimi K2.5 leads |
| FrontierMath v2 (Tier 4)Source | 4.200% | 2.083% | Kimi K2.5 leads |
| AA MATH-500Source | — | 99.2% | Not comparable |
Multilingual2 benchmarks
Multimodal6 benchmarks
Frequently Asked Questions (2)
Which is better, Kimi K2.5 or o3?
Kimi K2.5 is ahead on BenchLM's BenchAlign leaderboard, 59.66 to 47.89. The biggest single separator in this matchup is FrontierMath v2 (Tiers 1-3), where the scores are 27.900% and 18.685%.
Which is better for math, Kimi K2.5 or o3?
Kimi K2.5 has the edge for math in this comparison, averaging 60.6 versus 14.5. Inside this category, FrontierMath v2 (Tiers 1-3) 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
Choose a model with this week’s evidence
Join 2,000+ readers for ranking moves, pricing changes, and the claims that still need proof.
One email each week. Unsubscribe anytime.