Model comparison
DeepSeek-V2.5 (Sep 2024) vs Kimi K2 (Jul 2025)
Kimi K2 (Jul 2025) is the stronger model overall, scoring 41.2 to 37.6 on the Noometry Index.
Last verified . 18 shared benchmarks.
DeepSeekSummary
- They share 18 benchmarks with published results for both. DeepSeek-V2.5 (Sep 2024) scores higher in 1 category and Kimi K2 (Jul 2025) in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Kimi K2 (Jul 2025) leads 62.3 to 49.8.
- The biggest single-benchmark swing is Aider Polyglot: 17.8% for DeepSeek-V2.5 (Sep 2024) and 59.1% for Kimi K2 (Jul 2025).
Side by side
| DeepSeek-V2.5 (Sep 2024) | Kimi K2 (Jul 2025) | |
|---|---|---|
| Provider | DeepSeek | Moonshot AI |
| Noometry Index | 37.6 | 41.2 |
| Released | 2024-09-06 | 2025-07-12 |
| Weights | Open | Open |
| Context window | — | 262K |
| Max output | — | 262K |
| Input $ / M tokens | — | $0.57 |
| Output $ / M tokens | — | $2.30 |
| Results tracked | 22 | 42 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K2 (Jul 2025) leads
DeepSeek-V2.5 (Sep 2024): 31.7 (#281), Kimi K2 (Jul 2025): 42.4 (#102)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Kimi K2 (Jul 2025) |
|---|---|---|
| Aider Polyglot | 17.8% | 59.1% |
| LMArena Coding | 1309 | 1399 |
| SWE-bench Verified (bash only) | — | 63.4% |
| GSO | — | 4.9% |
| WeirdML | — | 42.8% |
| BigCodeBench Instruct | 48.6% | — |
| BigCodeBench Complete | 53.2% | — |
| ALE-Bench | — | 597.5 |
| HumanEval+ | 83.5% | — |
| MBPP+ | 74.1% | — |
Agentic & Tool Use Not comparable
DeepSeek-V2.5 (Sep 2024): —, Kimi K2 (Jul 2025): 32.4 (#64)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Kimi K2 (Jul 2025) |
|---|---|---|
| Terminal-Bench | — | 35.7% |
| Berkeley Function Calling Leaderboard | — | 59.1% |
| METR Time Horizons | — | 59.2% |
Reasoning DeepSeek-V2.5 (Sep 2024) leads
DeepSeek-V2.5 (Sep 2024): 25.6 (#145), Kimi K2 (Jul 2025): 23.3 (#179)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Hard Prompts | 1289 | 1384 |
| SimpleBench | — | 26.3% |
| Kagi LLM Benchmark | — | 64.4% |
| Epoch Capabilities Index | — | 146.01 |
| ForecastBench | — | 60.2 |
Math Kimi K2 (Jul 2025) leads
DeepSeek-V2.5 (Sep 2024): 35.9 (#177), Kimi K2 (Jul 2025): 42.7 (#83)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Math | 1288 | 1397 |
| Omni-MATH | — | 65.4% |
| FrontierMath (Feb 2025 set) | — | 21.4% |
| FrontierMath Tier 4 (v1) | — | 0% |
Knowledge Kimi K2 (Jul 2025) leads
DeepSeek-V2.5 (Sep 2024): 34.8 (#193), Kimi K2 (Jul 2025): 37.3 (#157)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Expert | 1266 | 1365 |
| MMLU-Pro | — | 81.9% |
| Confabulations | — | 20.4% |
| Vectara Hallucination Rate | — | 17.9% |
| GPQA (HELM) | — | 65.3% |
Multilingual Kimi K2 (Jul 2025) leads
DeepSeek-V2.5 (Sep 2024): 42.5 (#193), Kimi K2 (Jul 2025): 49.6 (#130)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Non-English | 1273 | 1372 |
| LMArena Chinese | 1318 | 1415 |
| LMArena French | 1289 | 1379 |
| LMArena German | 1258 | 1387 |
| LMArena Japanese | 1228 | 1349 |
| LMArena Korean | 1209 | 1325 |
| LMArena Russian | 1289 | 1385 |
| LMArena Spanish | 1248 | 1386 |
Instruction Following Kimi K2 (Jul 2025) leads
DeepSeek-V2.5 (Sep 2024): 67.5 (#194), Kimi K2 (Jul 2025): 71.1 (#156)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Instruction Following | 1280 | 1348 |
| IFEval | — | 85% |
Long Context Kimi K2 (Jul 2025) leads
DeepSeek-V2.5 (Sep 2024): 39.5 (#174), Kimi K2 (Jul 2025): 41.2 (#145)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Longer Query | 1301 | 1353 |
| Fiction.LiveBench | — | 66.7% |
| CL-bench | — | 17.6% |
Writing & Preference Kimi K2 (Jul 2025) leads
DeepSeek-V2.5 (Sep 2024): 49.8 (#187), Kimi K2 (Jul 2025): 62.3 (#78)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Text | 1294 | 1380 |
| LMArena Creative Writing | 1285 | 1350 |
| LMArena Multi-Turn | 1297 | 1371 |
| Short-Story Creative Writing | — | 85.6% |
| EQ-Bench Creative Writing | — | 1666 |
| WildBench | — | 86.2% |
Frequently asked questions
Is DeepSeek-V2.5 (Sep 2024) better than Kimi K2 (Jul 2025)?
Kimi K2 (Jul 2025) is the stronger model overall, scoring 41.2 to 37.6 on the Noometry Index.
Is DeepSeek-V2.5 (Sep 2024) or Kimi K2 (Jul 2025) better for coding?
Kimi K2 (Jul 2025) scores higher on coding benchmarks: 42.4 versus 31.7 in the Noometry coding category.
How many benchmarks do DeepSeek-V2.5 (Sep 2024) and Kimi K2 (Jul 2025) share?
18 benchmarks have published results for both models. DeepSeek-V2.5 (Sep 2024) has 22 scored results on Noometry and Kimi K2 (Jul 2025) has 42.