Model comparison
GPT-5.6 Luna vs Kimi K2 (Jul 2025)
GPT-5.6 Luna is the stronger model overall, scoring 54.6 to 41.2 on the Noometry Index.
Last verified . 23 shared benchmarks.
Summary
- They share 23 benchmarks with published results for both. GPT-5.6 Luna scores higher in 9 categories and Kimi K2 (Jul 2025) in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.6 Luna leads 77.7 to 42.7.
- The biggest single-benchmark swing is SimpleBench: 46.8% for GPT-5.6 Luna and 26.3% for Kimi K2 (Jul 2025).
- GPT-5.6 Luna is cheaper at $0.20 / $1.20 per million input/output tokens, against $0.57 / $2.30 for Kimi K2 (Jul 2025).
- GPT-5.6 Luna accepts more context: 1.05M tokens versus 262K.
- Kimi K2 (Jul 2025) has downloadable open weights; the other is API-only.
Side by side
| GPT-5.6 Luna | Kimi K2 (Jul 2025) | |
|---|---|---|
| Provider | OpenAI | Moonshot AI |
| Noometry Index | 54.6 | 41.2 |
| Released | 2026-07-09 | 2025-07-12 |
| Weights | Proprietary | Open |
| Context window | 1.05M | 262K |
| Max output | 128K | 262K |
| Input $ / M tokens | $0.20 | $0.57 |
| Output $ / M tokens | $1.20 | $2.30 |
| Results tracked | 52 | 42 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.6 Luna leads
GPT-5.6 Luna: 54.5 (#28), Kimi K2 (Jul 2025): 42.4 (#102)
| Benchmark | GPT-5.6 Luna | Kimi K2 (Jul 2025) |
|---|---|---|
| WeirdML | 60.9% | 42.8% |
| LMArena Coding | 1466 | 1399 |
| ALE-Bench | 1,667 | 597.5 |
| DeepSWE | 67.2% | — |
| FrontierCode | 39.8% | — |
| SWE-bench Verified (bash only) | — | 63.4% |
| Aider Polyglot | — | 59.1% |
| CursorBench | 35.9% | — |
| LMArena WebDev | 1519 | — |
| SciCode | 53.6% | — |
| GSO | — | 4.9% |
Agentic & Tool Use GPT-5.6 Luna leads
GPT-5.6 Luna: 34.4 (#45), Kimi K2 (Jul 2025): 32.4 (#64)
| Benchmark | GPT-5.6 Luna | Kimi K2 (Jul 2025) |
|---|---|---|
| Terminal-Bench | — | 35.7% |
| APEX-Agents | 43% | — |
| Berkeley Function Calling Leaderboard | — | 59.1% |
| BALROG | 45.6% | — |
| GDP.pdf | 22.7% | — |
| METR Time Horizons | — | 59.2% |
| Vending-Bench 2 | 4,095 | — |
Reasoning GPT-5.6 Luna leads
GPT-5.6 Luna: 47.6 (#43), Kimi K2 (Jul 2025): 23.3 (#179)
| Benchmark | GPT-5.6 Luna | Kimi K2 (Jul 2025) |
|---|---|---|
| SimpleBench | 46.8% | 26.3% |
| Kagi LLM Benchmark | 49.1% | 64.4% |
| LMArena Hard Prompts | 1451 | 1384 |
| Epoch Capabilities Index | 156.39 | 146.01 |
| ARC-AGI-2 | 59.5% | — |
| NYT Connections (extended) | 69.4% | — |
| ARC-AGI-1 | 88% | — |
| CritPt | 20.6% | — |
| Chess Puzzles | 40% | — |
| Mystery Game Puzzles | 21% | — |
| DTBench | 89.1% | — |
| LMCA | 48.5% | — |
| Surface Evolver Bench | 61.9% | — |
| ForecastBench | — | 60.2 |
Math GPT-5.6 Luna leads
GPT-5.6 Luna: 77.7 (#14), Kimi K2 (Jul 2025): 42.7 (#83)
| Benchmark | GPT-5.6 Luna | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Math | 1458 | 1397 |
| FrontierMath (Tiers 1-3) | 82.1% | — |
| FrontierMath Tier 4 | 61% | — |
| OTIS Mock AIME 2024-2025 | 98.3% | — |
| ProofBench | 60% | — |
| Omni-MATH | — | 65.4% |
| FrontierMath (Feb 2025 set) | — | 21.4% |
| FrontierMath Tier 4 (v1) | — | 0% |
Knowledge GPT-5.6 Luna leads
GPT-5.6 Luna: 58.5 (#34), Kimi K2 (Jul 2025): 37.3 (#157)
| Benchmark | GPT-5.6 Luna | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Expert | 1478 | 1365 |
| GPQA Diamond | 91.6% | — |
| SimpleQA Verified | 41% | — |
| MMLU-Pro | — | 81.9% |
| Confabulations | — | 20.4% |
| Vectara Hallucination Rate | — | 17.9% |
| GPQA (HELM) | — | 65.3% |
Multimodal Not comparable
GPT-5.6 Luna: 42.7 (#28), Kimi K2 (Jul 2025): —
| Benchmark | GPT-5.6 Luna | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Vision | 1258 | — |
| Blueprint-Bench 2 | 22.6% | — |
| Furniture Assembly | 42.5% | — |
| LMArena Document | 1457 | — |
Multilingual GPT-5.6 Luna leads
GPT-5.6 Luna: 52.8 (#78), Kimi K2 (Jul 2025): 49.6 (#130)
| Benchmark | GPT-5.6 Luna | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Non-English | 1417 | 1372 |
| LMArena Chinese | 1470 | 1415 |
| LMArena French | 1456 | 1379 |
| LMArena German | 1454 | 1387 |
| LMArena Japanese | 1411 | 1349 |
| LMArena Korean | 1415 | 1325 |
| LMArena Russian | 1428 | 1385 |
| LMArena Spanish | 1448 | 1386 |
Instruction Following GPT-5.6 Luna leads
GPT-5.6 Luna: 75.6 (#57), Kimi K2 (Jul 2025): 71.1 (#156)
| Benchmark | GPT-5.6 Luna | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Instruction Following | 1437 | 1348 |
| IFEval | — | 85% |
Long Context GPT-5.6 Luna leads
GPT-5.6 Luna: 43.9 (#82), Kimi K2 (Jul 2025): 41.2 (#145)
| Benchmark | GPT-5.6 Luna | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Longer Query | 1436 | 1353 |
| Fiction.LiveBench | — | 66.7% |
| CL-bench | — | 17.6% |
Writing & Preference GPT-5.6 Luna leads
GPT-5.6 Luna: 68.0 (#29), Kimi K2 (Jul 2025): 62.3 (#78)
| Benchmark | GPT-5.6 Luna | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Text | 1431 | 1380 |
| LMArena Creative Writing | 1396 | 1350 |
| EQ-Bench Creative Writing | 1829 | 1666 |
| LMArena Multi-Turn | 1434 | 1371 |
| Short-Story Creative Writing | — | 85.6% |
| WildBench | — | 86.2% |
| EQ-Bench 4 | 1156 | — |
Frequently asked questions
Is GPT-5.6 Luna better than Kimi K2 (Jul 2025)?
GPT-5.6 Luna is the stronger model overall, scoring 54.6 to 41.2 on the Noometry Index.
Which is cheaper, GPT-5.6 Luna or Kimi K2 (Jul 2025)?
GPT-5.6 Luna is cheaper. It lists at $0.20 per million input tokens and $1.20 per million output tokens; Kimi K2 (Jul 2025) lists at $0.57 and $2.30.
Is GPT-5.6 Luna or Kimi K2 (Jul 2025) better for coding?
GPT-5.6 Luna scores higher on coding benchmarks: 54.5 versus 42.4 in the Noometry coding category.
Which has the bigger context window?
GPT-5.6 Luna does, with 1.05M tokens against 262K.
How many benchmarks do GPT-5.6 Luna and Kimi K2 (Jul 2025) share?
23 benchmarks have published results for both models. GPT-5.6 Luna has 52 scored results on Noometry and Kimi K2 (Jul 2025) has 42.