Model comparison
Kimi K2.5 vs o1
Kimi K2.5 is the stronger model overall, scoring 47.8 to 40.6 on the Noometry Index.
Last verified . 30 shared benchmarks.
Summary
- They share 30 benchmarks with published results for both. Kimi K2.5 scores higher in 10 categories and o1 in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Kimi K2.5 leads 50.3 to 34.9.
- The biggest single-benchmark swing is ARC-AGI-1: 65.3% for Kimi K2.5 and 30.7% for o1.
- Kimi K2.5 is cheaper at $0.45 / $2.25 per million input/output tokens, against $15 / $60 for o1.
- Kimi K2.5 accepts more context: 262K tokens versus 200K.
- Kimi K2.5 has downloadable open weights; the other is API-only.
Side by side
| Kimi K2.5 | o1 | |
|---|---|---|
| Provider | Moonshot AI | OpenAI |
| Noometry Index | 47.8 | 40.6 |
| Released | 2026-01-27 | 2024-09-12 |
| Weights | Open | Proprietary |
| Context window | 262K | 200K |
| Max output | 262K | 100K |
| Input $ / M tokens | $0.45 | $15 |
| Output $ / M tokens | $2.25 | $60 |
| Results tracked | 51 | 52 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K2.5 leads
Kimi K2.5: 48.7 (#48), o1: 45.7 (#64)
| Benchmark | Kimi K2.5 | o1 |
|---|---|---|
| WeirdML | 45.6% | 47.6% |
| LMArena Coding | 1474 | 1367 |
| SWE-bench Verified | 73.8% | — |
| SWE-bench Verified (bash only) | 70.8% | — |
| Aider Polyglot | — | 61.7% |
| LMArena WebDev | 1436 | — |
| SWE-bench Multilingual | 67.3% | — |
| SciCode | 49% | — |
| LiveBench Coding | — | 69.7% |
| CadEval | — | 56% |
| ALE-Bench | 821.65 | — |
| HumanEval+ | — | 89% |
| MBPP+ | — | 80.2% |
Agentic & Tool Use Kimi K2.5 leads
Kimi K2.5: 34.0 (#44), o1: 24.3 (#111)
| Benchmark | Kimi K2.5 | o1 |
|---|---|---|
| Terminal-Bench | 43.2% | — |
| Cybench | — | 10% |
| OSWorld | 63.3% | — |
| METR Time Horizons | — | 51.1% |
| Vending-Bench 2 | 1,198 | — |
Reasoning Kimi K2.5 leads
Kimi K2.5: 31.2 (#74), o1: 27.5 (#103)
| Benchmark | Kimi K2.5 | o1 |
|---|---|---|
| SimpleBench | 46.8% | 41.7% |
| ARC-AGI-1 | 65.3% | 30.7% |
| Chess Puzzles | 12% | 15% |
| EnigmaEval | 3.4% | 5.7% |
| LMArena Hard Prompts | 1454 | 1372 |
| Epoch Capabilities Index | 148.05 | 141.92 |
| ARC-AGI-2 | 11.8% | — |
| Kagi LLM Benchmark | 78.5% | — |
| NYT Connections (extended) | 69.9% | — |
| CritPt | 3.1% | — |
| Thematic Generalization | 69.4% | — |
| LiveBench Reasoning | — | 91.6% |
| DTBench | — | 74.7% |
| LiveBench Data Analysis | — | 65.5% |
| LMCA | — | 22.3% |
| LiveBench | — | 75.7% |
Math Kimi K2.5 leads
Kimi K2.5: 50.3 (#46), o1: 34.9 (#150)
| Benchmark | Kimi K2.5 | o1 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 92.2% | 73.3% |
| LMArena Math | 1470 | 1388 |
| FrontierMath (Feb 2025 set) | 27.9% | 9.3% |
| FrontierMath (Tiers 1-3) | — | 14.7% |
| MathArena Final-Answer Competitions | 62.3% | — |
| LiveBench Math | — | 80.3% |
| MATH Level 5 | — | 94.7% |
| FrontierMath Tier 4 (v1) | 4.2% | — |
Knowledge Kimi K2.5 leads
Kimi K2.5: 53.6 (#52), o1: 41.5 (#101)
| Benchmark | Kimi K2.5 | o1 |
|---|---|---|
| GPQA Diamond | 87.6% | 76.8% |
| Humanity's Last Exam | 24.4% | 8% |
| SimpleQA Verified | 34.3% | 41.1% |
| LMArena Expert | 1466 | 1362 |
| Confabulations | — | 11.7% |
| Vectara Hallucination Rate | 14.2% | — |
Multimodal Kimi K2.5 leads
Kimi K2.5: 40.4 (#37), o1: 33.6 (#87)
| Benchmark | Kimi K2.5 | o1 |
|---|---|---|
| LMArena Vision | 1269 | 1168 |
| GeoBench | — | 80% |
| VPCT | — | 37% |
| LMArena Document | 1430 | — |
| SpatialViz-Bench | — | 41.4% |
Multilingual Kimi K2.5 leads
Kimi K2.5: 54.1 (#50), o1: 48.8 (#137)
| Benchmark | Kimi K2.5 | o1 |
|---|---|---|
| LMArena Non-English | 1433 | 1359 |
| LMArena Chinese | 1495 | 1394 |
| LMArena French | 1454 | 1343 |
| LMArena German | 1441 | 1337 |
| LMArena Japanese | 1422 | 1346 |
| LMArena Korean | 1412 | 1396 |
| LMArena Russian | 1436 | 1356 |
| LMArena Spanish | 1450 | 1345 |
Instruction Following Too close to call
Kimi K2.5: 75.5 (#60), o1: 75.0 (#82)
| Benchmark | Kimi K2.5 | o1 |
|---|---|---|
| LMArena Instruction Following | 1431 | 1368 |
| LiveBench Instruction Following | — | 81.5% |
Long Context Kimi K2.5 leads
Kimi K2.5: 52.1 (#7), o1: 50.3 (#9)
| Benchmark | Kimi K2.5 | o1 |
|---|---|---|
| Fiction.LiveBench | 86.1% | 83.3% |
| LMArena Longer Query | 1445 | 1378 |
| CL-bench | 19.3% | — |
| CL-bench Life | 13.2% | — |
Writing & Preference Kimi K2.5 leads
Kimi K2.5: 65.3 (#51), o1: 55.8 (#139)
| Benchmark | Kimi K2.5 | o1 |
|---|---|---|
| LMArena Text | 1445 | 1366 |
| LMArena Creative Writing | 1423 | 1347 |
| LMArena Multi-Turn | 1444 | 1368 |
| Short-Story Creative Writing | — | 70.2% |
| EQ-Bench Creative Writing | 1579 | — |
| LiveBench Language | — | 65.4% |
Frequently asked questions
Is Kimi K2.5 better than o1?
Kimi K2.5 is the stronger model overall, scoring 47.8 to 40.6 on the Noometry Index.
Which is cheaper, Kimi K2.5 or o1?
Kimi K2.5 is cheaper. It lists at $0.45 per million input tokens and $2.25 per million output tokens; o1 lists at $15 and $60.
Is Kimi K2.5 or o1 better for coding?
Kimi K2.5 scores higher on coding benchmarks: 48.7 versus 45.7 in the Noometry coding category.
Which has the bigger context window?
Kimi K2.5 does, with 262K tokens against 200K.
How many benchmarks do Kimi K2.5 and o1 share?
30 benchmarks have published results for both models. Kimi K2.5 has 51 scored results on Noometry and o1 has 52.