Model comparison
Kimi K2 (Jul 2025) vs MiniMax-M2.5
Kimi K2 (Jul 2025) is the stronger model overall, scoring 41.2 to 38.3 on the Noometry Index. MiniMax-M2.5 costs 1.9× less per token, which makes it the better buy when Kimi K2 (Jul 2025)'s lead doesn't matter for your workload.
Last verified . 25 shared benchmarks.
MiniMaxSummary
- They share 25 benchmarks with published results for both. Kimi K2 (Jul 2025) scores higher in 6 categories and MiniMax-M2.5 in 3 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Kimi K2 (Jul 2025) leads 42.7 to 26.9.
- The biggest single-benchmark swing is SWE-bench Verified (bash only): 63.4% for Kimi K2 (Jul 2025) and 75.8% for MiniMax-M2.5.
- MiniMax-M2.5 is cheaper at $0.30 / $1.20 per million input/output tokens, against $0.57 / $2.30 for Kimi K2 (Jul 2025).
- Kimi K2 (Jul 2025) accepts more context: 262K tokens versus 205K.
Side by side
| Kimi K2 (Jul 2025) | MiniMax-M2.5 | |
|---|---|---|
| Provider | Moonshot AI | MiniMax |
| Noometry Index | 41.2 | 38.3 |
| Released | 2025-07-12 | 2026-02-12 |
| Weights | Open | Open |
| Context window | 262K | 205K |
| Max output | 262K | 131K |
| Input $ / M tokens | $0.57 | $0.30 |
| Output $ / M tokens | $2.30 | $1.20 |
| Results tracked | 42 | 33 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding MiniMax-M2.5 leads
Kimi K2 (Jul 2025): 42.4 (#102), MiniMax-M2.5: 48.1 (#58)
| Benchmark | Kimi K2 (Jul 2025) | MiniMax-M2.5 |
|---|---|---|
| SWE-bench Verified (bash only) | 63.4% | 75.8% |
| LMArena Coding | 1399 | 1381 |
| ALE-Bench | 597.5 | 618.17 |
| Aider Polyglot | 59.1% | — |
| LMArena WebDev | — | 1387 |
| SWE-bench Multilingual | — | 68.3% |
| GSO | 4.9% | — |
| WeirdML | 42.8% | — |
Agentic & Tool Use Kimi K2 (Jul 2025) leads
Kimi K2 (Jul 2025): 32.4 (#64), MiniMax-M2.5: 30.4 (#77)
| Benchmark | Kimi K2 (Jul 2025) | MiniMax-M2.5 |
|---|---|---|
| Terminal-Bench | 35.7% | 42.7% |
| Berkeley Function Calling Leaderboard | 59.1% | — |
| METR Time Horizons | 59.2% | — |
| Vending-Bench 2 | — | -23.16 |
Reasoning Kimi K2 (Jul 2025) leads
Kimi K2 (Jul 2025): 23.3 (#179), MiniMax-M2.5: 17.5 (#292)
| Benchmark | Kimi K2 (Jul 2025) | MiniMax-M2.5 |
|---|---|---|
| Kagi LLM Benchmark | 64.4% | 55.2% |
| LMArena Hard Prompts | 1384 | 1372 |
| Epoch Capabilities Index | 146.01 | 146.68 |
| ARC-AGI-2 | — | 4.9% |
| SimpleBench | 26.3% | — |
| NYT Connections (extended) | — | 16.8% |
| ARC-AGI-1 | — | 63.7% |
| ForecastBench | 60.2 | — |
Math Kimi K2 (Jul 2025) leads
Kimi K2 (Jul 2025): 42.7 (#83), MiniMax-M2.5: 26.9 (#253)
| Benchmark | Kimi K2 (Jul 2025) | MiniMax-M2.5 |
|---|---|---|
| LMArena Math | 1397 | 1378 |
| ProofBench | — | 4% |
| Omni-MATH | 65.4% | — |
| FrontierMath (Feb 2025 set) | 21.4% | — |
| FrontierMath Tier 4 (v1) | 0% | — |
Knowledge MiniMax-M2.5 leads
Kimi K2 (Jul 2025): 37.3 (#157), MiniMax-M2.5: 39.2 (#135)
| Benchmark | Kimi K2 (Jul 2025) | MiniMax-M2.5 |
|---|---|---|
| Vectara Hallucination Rate | 17.9% | 9.1% |
| LMArena Expert | 1365 | 1379 |
| MMLU-Pro | 81.9% | — |
| Confabulations | 20.4% | — |
| GPQA (HELM) | 65.3% | — |
Multilingual Kimi K2 (Jul 2025) leads
Kimi K2 (Jul 2025): 49.6 (#130), MiniMax-M2.5: 47.1 (#152)
| Benchmark | Kimi K2 (Jul 2025) | MiniMax-M2.5 |
|---|---|---|
| LMArena Non-English | 1372 | 1338 |
| LMArena Chinese | 1415 | 1393 |
| LMArena French | 1379 | 1362 |
| LMArena German | 1387 | 1362 |
| LMArena Japanese | 1349 | 1171 |
| LMArena Korean | 1325 | 1232 |
| LMArena Russian | 1385 | 1358 |
| LMArena Spanish | 1386 | 1354 |
Instruction Following Too close to call
Kimi K2 (Jul 2025): 71.1 (#156), MiniMax-M2.5: 71.5 (#148)
| Benchmark | Kimi K2 (Jul 2025) | MiniMax-M2.5 |
|---|---|---|
| LMArena Instruction Following | 1348 | 1353 |
| IFEval | 85% | — |
Long Context Kimi K2 (Jul 2025) leads
Kimi K2 (Jul 2025): 41.2 (#145), MiniMax-M2.5: 37.5 (#216)
| Benchmark | Kimi K2 (Jul 2025) | MiniMax-M2.5 |
|---|---|---|
| CL-bench | 17.6% | 11.4% |
| LMArena Longer Query | 1353 | 1366 |
| Fiction.LiveBench | 66.7% | — |
| CL-bench Life | — | 6.3% |
Writing & Preference Kimi K2 (Jul 2025) leads
Kimi K2 (Jul 2025): 62.3 (#78), MiniMax-M2.5: 53.9 (#153)
| Benchmark | Kimi K2 (Jul 2025) | MiniMax-M2.5 |
|---|---|---|
| LMArena Text | 1380 | 1359 |
| LMArena Creative Writing | 1350 | 1331 |
| EQ-Bench Creative Writing | 1666 | 1361 |
| LMArena Multi-Turn | 1371 | 1364 |
| Short-Story Creative Writing | 85.6% | — |
| WildBench | 86.2% | — |
Frequently asked questions
Is Kimi K2 (Jul 2025) better than MiniMax-M2.5?
Kimi K2 (Jul 2025) is the stronger model overall, scoring 41.2 to 38.3 on the Noometry Index. MiniMax-M2.5 costs 1.9× less per token, which makes it the better buy when Kimi K2 (Jul 2025)'s lead doesn't matter for your workload.
Which is cheaper, Kimi K2 (Jul 2025) or MiniMax-M2.5?
MiniMax-M2.5 is cheaper. It lists at $0.30 per million input tokens and $1.20 per million output tokens; Kimi K2 (Jul 2025) lists at $0.57 and $2.30.
Is Kimi K2 (Jul 2025) or MiniMax-M2.5 better for coding?
MiniMax-M2.5 scores higher on coding benchmarks: 48.1 versus 42.4 in the Noometry coding category.
Which has the bigger context window?
Kimi K2 (Jul 2025) does, with 262K tokens against 205K.
How many benchmarks do Kimi K2 (Jul 2025) and MiniMax-M2.5 share?
25 benchmarks have published results for both models. Kimi K2 (Jul 2025) has 42 scored results on Noometry and MiniMax-M2.5 has 33.