Model comparison
Kimi K2.5 vs o4-mini
Kimi K2.5 is the stronger model overall, scoring 48.1 to 41.6 on the Noometry Index.
Last verified . 37 shared benchmarks.
Summary
- They share 37 benchmarks with published results for both. Kimi K2.5 scores higher in 10 categories and o4-mini in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Kimi K2.5 leads 65.1 to 54.0.
- The biggest single-benchmark swing is SWE-bench Verified (bash only): 70.8% for Kimi K2.5 and 45% for o4-mini.
- Kimi K2.5 is cheaper at $0.45 / $2.25 per million input/output tokens, against $1.10 / $4.40 for o4-mini.
- 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 | o4-mini | |
|---|---|---|
| Provider | Moonshot AI | OpenAI |
| Noometry Index | 48.1 | 41.6 |
| Released | 2026-01-27 | 2025-04-16 |
| Weights | Open | Proprietary |
| Context window | 262K | 200K |
| Max output | 262K | 100K |
| Input $ / M tokens | $0.45 | $1.10 |
| Output $ / M tokens | $2.25 | $4.40 |
| Results tracked | 51 | 60 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K2.5 leads
Kimi K2.5: 48.8 (#53), o4-mini: 40.9 (#127)
| Benchmark | Kimi K2.5 | o4-mini |
|---|---|---|
| SWE-bench Verified (bash only) | 70.8% | 45% |
| WeirdML | 45.6% | 52.6% |
| LMArena Coding | 1474 | 1368 |
| ALE-Bench | 821.65 | 826.17 |
| SWE-bench Verified | 73.8% | — |
| Aider Polyglot | — | 72% |
| LMArena WebDev | 1437 | — |
| SWE-bench Multilingual | 67.3% | — |
| SciCode | 49% | — |
| GSO | — | 3.6% |
| CadEval | — | 62% |
| AlgoTune | — | 1.72 |
Agentic & Tool Use Kimi K2.5 leads
Kimi K2.5: 34.2 (#48), o4-mini: 32.6 (#61)
| Benchmark | Kimi K2.5 | o4-mini |
|---|---|---|
| Terminal-Bench | 43.2% | — |
| Berkeley Function Calling Leaderboard | — | 53.2% |
| GDPval | — | 25.3% |
| OSWorld | 63.3% | — |
| METR Time Horizons | — | 63.9% |
| Vending-Bench 2 | 1,198 | — |
Reasoning Kimi K2.5 leads
Kimi K2.5: 31.2 (#80), o4-mini: 24.6 (#162)
| Benchmark | Kimi K2.5 | o4-mini |
|---|---|---|
| ARC-AGI-2 | 11.8% | 6.1% |
| SimpleBench | 46.8% | 38.7% |
| Kagi LLM Benchmark | 78.5% | 67.6% |
| ARC-AGI-1 | 65.3% | 58.7% |
| CritPt | 3.1% | 0.6% |
| Chess Puzzles | 12% | 26% |
| EnigmaEval | 3.4% | 9.2% |
| LMArena Hard Prompts | 1453 | 1351 |
| Epoch Capabilities Index | 148.03 | 145.64 |
| NYT Connections (extended) | 69.9% | — |
| Thematic Generalization | 69.4% | — |
| Mystery Game Puzzles | — | 5% |
| DTBench | — | 77.6% |
| LMCA | — | 26.5% |
| ForecastBench | — | 61.8 |
Math Kimi K2.5 leads
Kimi K2.5: 51.8 (#53), o4-mini: 40.8 (#89)
| Benchmark | Kimi K2.5 | o4-mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 92.2% | 81.7% |
| LMArena Math | 1470 | 1389 |
| FrontierMath (Feb 2025 set) | 27.9% | 24.8% |
| FrontierMath Tier 4 (v1) | 4.2% | 6.3% |
| FrontierMath (Tiers 1-3) | — | 36.1% |
| FrontierMath Tier 4 | — | 4.9% |
| MathArena Final-Answer Competitions | 62.3% | — |
| Omni-MATH | — | 72% |
| MATH Level 5 | — | 97.8% |
Knowledge Kimi K2.5 leads
Kimi K2.5: 53.6 (#56), o4-mini: 43.6 (#91)
| Benchmark | Kimi K2.5 | o4-mini |
|---|---|---|
| GPQA Diamond | 87.6% | 79.6% |
| Humanity's Last Exam | 24.4% | 18.1% |
| SimpleQA Verified | 34.3% | 19.6% |
| Vectara Hallucination Rate | 14.2% | 18.6% |
| LMArena Expert | 1466 | 1343 |
| MMLU-Pro | — | 82% |
| Confabulations | — | 15.8% |
| GPQA (HELM) | — | 73.5% |
Multimodal Too close to call
Kimi K2.5: 41.1 (#39), o4-mini: 40.2 (#49)
| Benchmark | Kimi K2.5 | o4-mini |
|---|---|---|
| LMArena Vision | 1269 | 1194 |
| GeoBench | — | 64% |
| VPCT | — | 57.5% |
| LMArena Document | 1430 | — |
Multilingual Kimi K2.5 leads
Kimi K2.5: 53.9 (#53), o4-mini: 47.0 (#154)
| Benchmark | Kimi K2.5 | o4-mini |
|---|---|---|
| LMArena Non-English | 1433 | 1337 |
| LMArena Chinese | 1495 | 1354 |
| LMArena French | 1454 | 1364 |
| LMArena German | 1441 | 1336 |
| LMArena Japanese | 1421 | 1308 |
| LMArena Korean | 1410 | 1312 |
| LMArena Russian | 1435 | 1334 |
| LMArena Spanish | 1450 | 1347 |
Instruction Following Too close to call
Kimi K2.5: 75.3 (#64), o4-mini: 75.2 (#68)
| Benchmark | Kimi K2.5 | o4-mini |
|---|---|---|
| LMArena Instruction Following | 1431 | 1321 |
| IFEval | — | 92.8% |
Long Context Kimi K2.5 leads
Kimi K2.5: 52.1 (#7), o4-mini: 45.5 (#33)
| Benchmark | Kimi K2.5 | o4-mini |
|---|---|---|
| Fiction.LiveBench | 86.1% | 77.8% |
| LMArena Longer Query | 1445 | 1315 |
| CL-bench | 19.3% | — |
| CL-bench Life | 13.2% | — |
Writing & Preference Kimi K2.5 leads
Kimi K2.5: 65.1 (#53), o4-mini: 54.0 (#152)
| Benchmark | Kimi K2.5 | o4-mini |
|---|---|---|
| LMArena Text | 1445 | 1353 |
| LMArena Creative Writing | 1423 | 1294 |
| LMArena Multi-Turn | 1444 | 1350 |
| Short-Story Creative Writing | — | 75% |
| EQ-Bench Creative Writing | 1579 | — |
| WildBench | — | 85.4% |
Frequently asked questions
Is Kimi K2.5 better than o4-mini?
Kimi K2.5 is the stronger model overall, scoring 48.1 to 41.6 on the Noometry Index.
Which is cheaper, Kimi K2.5 or o4-mini?
Kimi K2.5 is cheaper. It lists at $0.45 per million input tokens and $2.25 per million output tokens; o4-mini lists at $1.10 and $4.40.
Is Kimi K2.5 or o4-mini better for coding?
Kimi K2.5 scores higher on coding benchmarks: 48.8 versus 40.9 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 o4-mini share?
37 benchmarks have published results for both models. Kimi K2.5 has 51 scored results on Noometry and o4-mini has 60.