Model comparison
GPT-6 Luna vs Kimi K2 (Jul 2025)
GPT-6 Luna is the stronger model overall, scoring 53.3 to 41.2 on the Noometry Index.
Last verified . 19 shared benchmarks.
Summary
- They share 19 benchmarks with published results for both. GPT-6 Luna scores higher in 8 categories and Kimi K2 (Jul 2025) in 1 category; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-6 Luna leads 76.1 to 42.7.
- GPT-6 Luna is cheaper at $0.10 / $0.50 per million input/output tokens, against $0.57 / $2.30 for Kimi K2 (Jul 2025).
- GPT-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-6 Luna | Kimi K2 (Jul 2025) | |
|---|---|---|
| Provider | OpenAI | Moonshot AI |
| Noometry Index | 53.3 | 41.2 |
| Released | 2026-09-22 | 2025-07-12 |
| Weights | Proprietary | Open |
| Context window | 1.05M | 262K |
| Max output | 128K | 262K |
| Input $ / M tokens | $0.10 | $0.57 |
| Output $ / M tokens | $0.50 | $2.30 |
| Results tracked | 42 | 42 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-6 Luna leads
GPT-6 Luna: 55.5 (#25), Kimi K2 (Jul 2025): 42.4 (#102)
| Benchmark | GPT-6 Luna | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Coding | 1439 | 1399 |
| ALE-Bench | 1,577 | 597.5 |
| DeepSWE | 66.6% | — |
| FrontierCode | 42.4% | — |
| SWE-bench Verified (bash only) | — | 63.4% |
| Aider Polyglot | — | 59.1% |
| LMArena WebDev | 1581 | — |
| SciCode | 54.6% | — |
| GSO | — | 4.9% |
| WeirdML | — | 42.8% |
Agentic & Tool Use Too close to call
GPT-6 Luna: 33.3 (#54), Kimi K2 (Jul 2025): 32.4 (#64)
| Benchmark | GPT-6 Luna | Kimi K2 (Jul 2025) |
|---|---|---|
| Terminal-Bench | — | 35.7% |
| APEX-Agents | 44.3% | — |
| Berkeley Function Calling Leaderboard | — | 59.1% |
| GDP.pdf | 23% | — |
| METR Time Horizons | — | 59.2% |
Reasoning GPT-6 Luna leads
GPT-6 Luna: 48.2 (#41), Kimi K2 (Jul 2025): 23.3 (#179)
| Benchmark | GPT-6 Luna | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Hard Prompts | 1411 | 1384 |
| Epoch Capabilities Index | 156.28 | 146.01 |
| ARC-AGI-2 | 59.3% | — |
| SimpleBench | — | 26.3% |
| Kagi LLM Benchmark | — | 64.4% |
| NYT Connections (extended) | 68.7% | — |
| ARC-AGI-1 | 86.7% | — |
| CritPt | 19.4% | — |
| Chess Puzzles | 31% | — |
| Mystery Game Puzzles | 7% | — |
| DTBench | 90.1% | — |
| LMCA | 44.5% | — |
| ForecastBench | — | 60.2 |
Math GPT-6 Luna leads
GPT-6 Luna: 76.1 (#15), Kimi K2 (Jul 2025): 42.7 (#83)
| Benchmark | GPT-6 Luna | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Math | 1416 | 1397 |
| FrontierMath (Tiers 1-3) | 78.9% | — |
| FrontierMath Tier 4 | 56.1% | — |
| OTIS Mock AIME 2024-2025 | 98.9% | — |
| ProofBench | 64% | — |
| Omni-MATH | — | 65.4% |
| FrontierMath (Feb 2025 set) | — | 21.4% |
| FrontierMath Tier 4 (v1) | — | 0% |
Knowledge GPT-6 Luna leads
GPT-6 Luna: 57.0 (#41), Kimi K2 (Jul 2025): 37.3 (#157)
| Benchmark | GPT-6 Luna | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Expert | 1444 | 1365 |
| GPQA Diamond | 90.5% | — |
| SimpleQA Verified | 41.4% | — |
| MMLU-Pro | — | 81.9% |
| Confabulations | — | 20.4% |
| Vectara Hallucination Rate | — | 17.9% |
| GPQA (HELM) | — | 65.3% |
Multimodal Not comparable
GPT-6 Luna: 42.4 (#30), Kimi K2 (Jul 2025): —
| Benchmark | GPT-6 Luna | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Vision | 1217 | — |
| Blueprint-Bench 2 | 31.2% | — |
| Furniture Assembly | 44.2% | — |
Multilingual Too close to call
GPT-6 Luna: 50.5 (#117), Kimi K2 (Jul 2025): 49.6 (#130)
| Benchmark | GPT-6 Luna | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Non-English | 1386 | 1372 |
| LMArena Chinese | 1433 | 1415 |
| LMArena French | 1420 | 1379 |
| LMArena German | 1369 | 1387 |
| LMArena Japanese | 1369 | 1349 |
| LMArena Korean | 1360 | 1325 |
| LMArena Russian | 1394 | 1385 |
| LMArena Spanish | 1393 | 1386 |
Instruction Following GPT-6 Luna leads
GPT-6 Luna: 74.3 (#99), Kimi K2 (Jul 2025): 71.1 (#156)
| Benchmark | GPT-6 Luna | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Instruction Following | 1409 | 1348 |
| IFEval | — | 85% |
Long Context GPT-6 Luna leads
GPT-6 Luna: 43.0 (#111), Kimi K2 (Jul 2025): 41.2 (#145)
| Benchmark | GPT-6 Luna | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Longer Query | 1409 | 1353 |
| Fiction.LiveBench | — | 66.7% |
| CL-bench | — | 17.6% |
Writing & Preference Kimi K2 (Jul 2025) leads
GPT-6 Luna: 58.3 (#119), Kimi K2 (Jul 2025): 62.3 (#78)
| Benchmark | GPT-6 Luna | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Text | 1391 | 1380 |
| LMArena Creative Writing | 1363 | 1350 |
| LMArena Multi-Turn | 1396 | 1371 |
| Short-Story Creative Writing | — | 85.6% |
| EQ-Bench Creative Writing | — | 1666 |
| WildBench | — | 86.2% |
Frequently asked questions
Is GPT-6 Luna better than Kimi K2 (Jul 2025)?
GPT-6 Luna is the stronger model overall, scoring 53.3 to 41.2 on the Noometry Index.
Which is cheaper, GPT-6 Luna or Kimi K2 (Jul 2025)?
GPT-6 Luna is cheaper. It lists at $0.10 per million input tokens and $0.50 per million output tokens; Kimi K2 (Jul 2025) lists at $0.57 and $2.30.
Is GPT-6 Luna or Kimi K2 (Jul 2025) better for coding?
GPT-6 Luna scores higher on coding benchmarks: 55.5 versus 42.4 in the Noometry coding category.
Which has the bigger context window?
GPT-6 Luna does, with 1.05M tokens against 262K.
How many benchmarks do GPT-6 Luna and Kimi K2 (Jul 2025) share?
19 benchmarks have published results for both models. GPT-6 Luna has 42 scored results on Noometry and Kimi K2 (Jul 2025) has 42.