Model comparison
Kimi K2.5 vs Kimi K2.7 Code
Kimi K2.5 is the stronger model overall, scoring 48.1 to 43.3 on the Noometry Index.
Last verified . 12 shared benchmarks.
Summary
- They share 12 benchmarks with published results for both. Kimi K2.5 scores higher in 3 categories and Kimi K2.7 Code in 2 categories; 4 gaps are clear of the uncertainty.
- The widest gap is in agentic & tool use, where Kimi K2.5 leads 34.2 to 24.0.
- The biggest single-benchmark swing is SimpleBench: 46.8% for Kimi K2.5 and 57.9% for Kimi K2.7 Code.
- Kimi K2.5 is cheaper at $0.45 / $2.25 per million input/output tokens, against $0.95 / $4 for Kimi K2.7 Code.
Side by side
| Kimi K2.5 | Kimi K2.7 Code | |
|---|---|---|
| Provider | Moonshot AI | Moonshot AI |
| Noometry Index | 48.1 | 43.3 |
| Released | 2026-01-27 | 2026-06-12 |
| Weights | Open | Open |
| Context window | 262K | 262K |
| Max output | 262K | 262K |
| Input $ / M tokens | $0.45 | $0.95 |
| Output $ / M tokens | $2.25 | $4 |
| Results tracked | 51 | 19 |
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), Kimi K2.7 Code: 42.9 (#95)
| Benchmark | Kimi K2.5 | Kimi K2.7 Code |
|---|---|---|
| LMArena WebDev | 1437 | 1473 |
| SciCode | 49% | 47.5% |
| WeirdML | 45.6% | 54.1% |
| ALE-Bench | 821.65 | 886.23 |
| SWE-bench Verified | 73.8% | — |
| DeepSWE | — | 30.5% |
| FrontierCode | — | 30.1% |
| SWE-bench Verified (bash only) | 70.8% | — |
| SWE-bench Multilingual | 67.3% | — |
| LMArena Coding | 1474 | — |
Agentic & Tool Use Kimi K2.5 leads
Kimi K2.5: 34.2 (#48), Kimi K2.7 Code: 24.0 (#122)
| Benchmark | Kimi K2.5 | Kimi K2.7 Code |
|---|---|---|
| Vending-Bench 2 | 1,198 | 5,083 |
| Terminal-Bench | 43.2% | — |
| APEX-Agents | — | 37.6% |
| OSWorld | 63.3% | — |
| GBAEval | — | 0.9% |
Reasoning Kimi K2.7 Code leads
Kimi K2.5: 31.2 (#80), Kimi K2.7 Code: 39.0 (#61)
| Benchmark | Kimi K2.5 | Kimi K2.7 Code |
|---|---|---|
| SimpleBench | 46.8% | 57.9% |
| CritPt | 3.1% | 10% |
| Chess Puzzles | 12% | 21% |
| Epoch Capabilities Index | 148.03 | 149.97 |
| ARC-AGI-2 | 11.8% | — |
| Kagi LLM Benchmark | 78.5% | — |
| NYT Connections (extended) | 69.9% | — |
| ARC-AGI-1 | 65.3% | — |
| EnigmaEval | 3.4% | — |
| Thematic Generalization | 69.4% | — |
| LMArena Hard Prompts | 1453 | — |
| Surface Evolver Bench | — | 48.8% |
Math Kimi K2.7 Code leads
Kimi K2.5: 51.8 (#53), Kimi K2.7 Code: 52.9 (#48)
| Benchmark | Kimi K2.5 | Kimi K2.7 Code |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 92.2% | 95.6% |
| FrontierMath (Tiers 1-3) | — | 54% |
| FrontierMath Tier 4 | — | 12.2% |
| MathArena Final-Answer Competitions | 62.3% | — |
| LMArena Math | 1470 | — |
| FrontierMath (Feb 2025 set) | 27.9% | — |
| FrontierMath Tier 4 (v1) | 4.2% | — |
Knowledge Too close to call
Kimi K2.5: 53.6 (#56), Kimi K2.7 Code: 53.5 (#57)
| Benchmark | Kimi K2.5 | Kimi K2.7 Code |
|---|---|---|
| GPQA Diamond | 87.6% | 87.9% |
| SimpleQA Verified | 34.3% | 36.5% |
| Humanity's Last Exam | 24.4% | — |
| Vectara Hallucination Rate | 14.2% | — |
| LMArena Expert | 1466 | — |
Multimodal Not comparable
Kimi K2.5: 41.1 (#39), Kimi K2.7 Code: —
| Benchmark | Kimi K2.5 | Kimi K2.7 Code |
|---|---|---|
| LMArena Vision | 1269 | — |
| LMArena Document | 1430 | — |
Multilingual Not comparable
Kimi K2.5: 53.9 (#53), Kimi K2.7 Code: —
| Benchmark | Kimi K2.5 | Kimi K2.7 Code |
|---|---|---|
| LMArena Non-English | 1433 | — |
| LMArena Chinese | 1495 | — |
| LMArena French | 1454 | — |
| LMArena German | 1441 | — |
| LMArena Japanese | 1421 | — |
| LMArena Korean | 1410 | — |
| LMArena Russian | 1435 | — |
| LMArena Spanish | 1450 | — |
Instruction Following Not comparable
Kimi K2.5: 75.3 (#64), Kimi K2.7 Code: —
| Benchmark | Kimi K2.5 | Kimi K2.7 Code |
|---|---|---|
| LMArena Instruction Following | 1431 | — |
Long Context Not comparable
Kimi K2.5: 52.1 (#7), Kimi K2.7 Code: —
| Benchmark | Kimi K2.5 | Kimi K2.7 Code |
|---|---|---|
| Fiction.LiveBench | 86.1% | — |
| CL-bench | 19.3% | — |
| CL-bench Life | 13.2% | — |
| LMArena Longer Query | 1445 | — |
Writing & Preference Not comparable
Kimi K2.5: 65.1 (#53), Kimi K2.7 Code: —
| Benchmark | Kimi K2.5 | Kimi K2.7 Code |
|---|---|---|
| LMArena Text | 1445 | — |
| LMArena Creative Writing | 1423 | — |
| EQ-Bench Creative Writing | 1579 | — |
| LMArena Multi-Turn | 1444 | — |
Frequently asked questions
Is Kimi K2.5 better than Kimi K2.7 Code?
Kimi K2.5 is the stronger model overall, scoring 48.1 to 43.3 on the Noometry Index.
Which is cheaper, Kimi K2.5 or Kimi K2.7 Code?
Kimi K2.5 is cheaper. It lists at $0.45 per million input tokens and $2.25 per million output tokens; Kimi K2.7 Code lists at $0.95 and $4.
Is Kimi K2.5 or Kimi K2.7 Code better for coding?
Kimi K2.5 scores higher on coding benchmarks: 48.8 versus 42.9 in the Noometry coding category.
Which has the bigger context window?
Both accept 262K tokens.
How many benchmarks do Kimi K2.5 and Kimi K2.7 Code share?
12 benchmarks have published results for both models. Kimi K2.5 has 51 scored results on Noometry and Kimi K2.7 Code has 19.