Model comparison
gpt-oss-120b vs Kimi K3
Kimi K3 is the stronger model overall, scoring 59.5 to 36.3 on the Noometry Index. gpt-oss-120b costs 85× less per token, which makes it the better buy when Kimi K3's lead doesn't matter for your workload.
Last verified . 33 shared benchmarks.
Summary
- They share 33 benchmarks with published results for both. gpt-oss-120b scores higher in 0 categories and Kimi K3 in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Kimi K3 leads 63.0 to 20.0.
- The biggest single-benchmark swing is Surface Evolver Bench: 25% for gpt-oss-120b and 95% for Kimi K3.
- gpt-oss-120b is cheaper at $0.037 / $0.17 per million input/output tokens, against $3 / $15 for Kimi K3.
- Kimi K3 accepts more context: 1.05M tokens versus 131K.
Side by side
| gpt-oss-120b | Kimi K3 | |
|---|---|---|
| Provider | OpenAI | Moonshot AI |
| Noometry Index | 36.3 | 59.5 |
| Released | 2025-08-05 | 2026-07-16 |
| Weights | Open | Open |
| Context window | 131K | 1.05M |
| Max output | 41K | 1.05M |
| Input $ / M tokens | $0.037 | $3 |
| Output $ / M tokens | $0.17 | $15 |
| Results tracked | 48 | 53 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K3 leads
gpt-oss-120b: 33.5 (#256), Kimi K3: 61.0 (#10)
| Benchmark | gpt-oss-120b | Kimi K3 |
|---|---|---|
| SciCode | 36% | 59.5% |
| WeirdML | 48.2% | 82.6% |
| LMArena Coding | 1380 | 1508 |
| ALE-Bench | 575.62 | 1,524 |
| DeepSWE | — | 68.5% |
| FrontierCode | — | 44.2% |
| SWE-bench Verified (bash only) | 26% | — |
| Aider Polyglot | 41.8% | — |
| LMArena WebDev | — | 1654 |
| FrontierSWE | — | 25.9% |
| AlgoTune | 1.41 | — |
Agentic & Tool Use Kimi K3 leads
gpt-oss-120b: 12.2 (#153), Kimi K3: 41.8 (#20)
| Benchmark | gpt-oss-120b | Kimi K3 |
|---|---|---|
| APEX-Agents | 4.4% | 50.6% |
| Vending-Bench 2 | -21.53 | 5,165 |
| Terminal-Bench | 18.7% | — |
| τ²-bench Banking | — | 37.1% |
| PostTrainBench | — | 32% |
| GBAEval | — | 48.3% |
| GDP.pdf | — | 19% |
| METR Time Horizons | 56.6% | — |
Reasoning Kimi K3 leads
gpt-oss-120b: 20.0 (#245), Kimi K3: 63.0 (#17)
| Benchmark | gpt-oss-120b | Kimi K3 |
|---|---|---|
| SimpleBench | 22.1% | 60.7% |
| CritPt | 1.1% | 23.4% |
| Chess Puzzles | 20% | 39% |
| LMArena Hard Prompts | 1364 | 1496 |
| Mystery Game Puzzles | 2% | 26% |
| DTBench | 76.3% | 91.2% |
| LMCA | 22.1% | 52.7% |
| Surface Evolver Bench | 25% | 95% |
| Epoch Capabilities Index | 139.93 | 157.45 |
| ARC-AGI-2 | — | 60.4% |
| Kagi LLM Benchmark | 58.6% | — |
| NYT Connections (extended) | — | 93.6% |
| ARC-AGI-1 | — | 94.5% |
| ForecastBench | — | 61.1 |
Math Kimi K3 leads
gpt-oss-120b: 52.5 (#50), Kimi K3: 74.2 (#16)
| Benchmark | gpt-oss-120b | Kimi K3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 88.9% | 97.2% |
| LMArena Math | 1389 | 1491 |
| FrontierMath (Tiers 1-3) | — | 72.2% |
| FrontierMath Tier 4 | — | 39% |
| MathArena Final-Answer Competitions | — | 87.8% |
| ProofBench | — | 87% |
| Omni-MATH | 68.8% | — |
Knowledge Kimi K3 leads
gpt-oss-120b: 42.4 (#96), Kimi K3: 63.2 (#21)
| Benchmark | gpt-oss-120b | Kimi K3 |
|---|---|---|
| GPQA Diamond | 75.8% | 93.1% |
| LMArena Expert | 1356 | 1521 |
| SimpleQA Verified | — | 50.6% |
| MMLU-Pro | 79.5% | — |
| Confabulations | 15.7% | — |
| Vectara Hallucination Rate | 14.2% | — |
| GPQA (HELM) | 68.4% | — |
Multimodal Not comparable
gpt-oss-120b: —, Kimi K3: 37.8 (#70)
| Benchmark | gpt-oss-120b | Kimi K3 |
|---|---|---|
| Blueprint-Bench 2 | — | 29.5% |
| Furniture Assembly | — | 34.2% |
Multilingual Kimi K3 leads
gpt-oss-120b: 48.0 (#147), Kimi K3: 56.3 (#21)
| Benchmark | gpt-oss-120b | Kimi K3 |
|---|---|---|
| LMArena Non-English | 1351 | 1466 |
| LMArena Chinese | 1385 | 1529 |
| LMArena French | 1369 | 1491 |
| LMArena German | 1353 | 1488 |
| LMArena Japanese | 1331 | 1487 |
| LMArena Korean | 1282 | 1458 |
| LMArena Russian | 1343 | 1482 |
| LMArena Spanish | 1389 | 1472 |
Instruction Following Kimi K3 leads
gpt-oss-120b: 69.3 (#173), Kimi K3: 77.7 (#14)
| Benchmark | gpt-oss-120b | Kimi K3 |
|---|---|---|
| LMArena Instruction Following | 1318 | 1483 |
| IFEval | 83.6% | — |
Long Context Kimi K3 leads
gpt-oss-120b: 31.4 (#278), Kimi K3: 45.8 (#29)
| Benchmark | gpt-oss-120b | Kimi K3 |
|---|---|---|
| LMArena Longer Query | 1319 | 1494 |
| Fiction.LiveBench | 44.4% | — |
Writing & Preference Kimi K3 leads
gpt-oss-120b: 46.5 (#217), Kimi K3: 76.6 (#4)
| Benchmark | gpt-oss-120b | Kimi K3 |
|---|---|---|
| LMArena Text | 1365 | 1476 |
| LMArena Creative Writing | 1275 | 1454 |
| EQ-Bench Creative Writing | 961 | 2082 |
| LMArena Multi-Turn | 1340 | 1488 |
| Short-Story Creative Writing | 77.1% | — |
| WildBench | 84.5% | — |
| EQ-Bench 4 | — | 1339 |
Frequently asked questions
Is gpt-oss-120b better than Kimi K3?
Kimi K3 is the stronger model overall, scoring 59.5 to 36.3 on the Noometry Index. gpt-oss-120b costs 85× less per token, which makes it the better buy when Kimi K3's lead doesn't matter for your workload.
Which is cheaper, gpt-oss-120b or Kimi K3?
gpt-oss-120b is cheaper. It lists at $0.037 per million input tokens and $0.17 per million output tokens; Kimi K3 lists at $3 and $15.
Is gpt-oss-120b or Kimi K3 better for coding?
Kimi K3 scores higher on coding benchmarks: 61.0 versus 33.5 in the Noometry coding category.
Which has the bigger context window?
Kimi K3 does, with 1.05M tokens against 131K.
How many benchmarks do gpt-oss-120b and Kimi K3 share?
33 benchmarks have published results for both models. gpt-oss-120b has 48 scored results on Noometry and Kimi K3 has 53.