Model comparison
GPT-5.4 mini vs Kimi K2 Thinking Turbo
GPT-5.4 mini and Kimi K2 Thinking Turbo score almost the same on the Noometry Index (44.8 vs 45.4), so choose on price, context window or the category you care about most.
Last verified . 21 shared benchmarks.
Summary
- They share 21 benchmarks with published results for both. GPT-5.4 mini scores higher in 5 categories and Kimi K2 Thinking Turbo in 3 categories; 3 gaps are clear of the uncertainty.
- The widest gap is in coding, where GPT-5.4 mini leads 45.0 to 38.1.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 88.9% for GPT-5.4 mini and 83.1% for Kimi K2 Thinking Turbo.
- Kimi K2 Thinking Turbo has downloadable open weights; the other is API-only.
Side by side
| GPT-5.4 mini | Kimi K2 Thinking Turbo | |
|---|---|---|
| Provider | OpenAI | Moonshot AI |
| Noometry Index | 44.8 | 45.4 |
| Released | 2026-03-17 | 2025-11-06 |
| Weights | Proprietary | Open |
| Context window | 400K | — |
| Max output | 128K | — |
| Input $ / M tokens | $0.75 | — |
| Output $ / M tokens | $4.50 | — |
| Results tracked | 46 | 21 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.4 mini leads
GPT-5.4 mini: 45.0 (#67), Kimi K2 Thinking Turbo: 38.1 (#166)
| Benchmark | GPT-5.4 mini | Kimi K2 Thinking Turbo |
|---|---|---|
| LMArena WebDev | 1397 | 1322 |
| LMArena Coding | 1438 | 1454 |
| FrontierCode | 27% | — |
| SciCode | 49.9% | — |
| WeirdML | 60.3% | — |
| ALE-Bench | 1,189 | — |
Agentic & Tool Use Not comparable
GPT-5.4 mini: 29.6 (#74), Kimi K2 Thinking Turbo: —
| Benchmark | GPT-5.4 mini | Kimi K2 Thinking Turbo |
|---|---|---|
| DeepResearch Bench | 36.3% | — |
Reasoning Kimi K2 Thinking Turbo leads
GPT-5.4 mini: 30.3 (#77), Kimi K2 Thinking Turbo: 31.4 (#72)
| Benchmark | GPT-5.4 mini | Kimi K2 Thinking Turbo |
|---|---|---|
| Chess Puzzles | 24% | 20% |
| LMArena Hard Prompts | 1424 | 1428 |
| ARC-AGI-2 | 18.9% | — |
| Kagi LLM Benchmark | 37.9% | — |
| NYT Connections (extended) | 61.8% | — |
| ARC-AGI-1 | 63.7% | — |
| CritPt | 10% | — |
| Thematic Generalization | 61.7% | — |
| Mystery Game Puzzles | 11% | — |
| DTBench | 80% | — |
| LMCA | 40.8% | — |
| Epoch Capabilities Index | 148.83 | — |
| ForecastBench | 57 | — |
Math Too close to call
GPT-5.4 mini: 45.4 (#63), Kimi K2 Thinking Turbo: 45.5 (#62)
| Benchmark | GPT-5.4 mini | Kimi K2 Thinking Turbo |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 88.9% | 83.1% |
| LMArena Math | 1418 | 1429 |
| FrontierMath (Tiers 1-3) | 51.2% | — |
| FrontierMath Tier 4 | 9.8% | — |
| ProofBench | 21% | — |
| FrontierMath (Feb 2025 set) | 28.3% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge Too close to call
GPT-5.4 mini: 51.4 (#62), Kimi K2 Thinking Turbo: 50.9 (#64)
| Benchmark | GPT-5.4 mini | Kimi K2 Thinking Turbo |
|---|---|---|
| GPQA Diamond | 86.9% | 84.2% |
| LMArena Expert | 1434 | 1439 |
| SimpleQA Verified | 29.4% | — |
| Vectara Hallucination Rate | 5.5% | — |
Multimodal Not comparable
GPT-5.4 mini: 39.0 (#51), Kimi K2 Thinking Turbo: —
| Benchmark | GPT-5.4 mini | Kimi K2 Thinking Turbo |
|---|---|---|
| LMArena Vision | 1245 | — |
Multilingual Too close to call
GPT-5.4 mini: 52.1 (#94), Kimi K2 Thinking Turbo: 51.6 (#104)
| Benchmark | GPT-5.4 mini | Kimi K2 Thinking Turbo |
|---|---|---|
| LMArena Non-English | 1405 | 1398 |
| LMArena Chinese | 1446 | 1457 |
| LMArena French | 1440 | 1425 |
| LMArena German | 1407 | 1390 |
| LMArena Japanese | 1375 | 1356 |
| LMArena Korean | 1369 | 1329 |
| LMArena Russian | 1417 | 1390 |
| LMArena Spanish | 1403 | 1407 |
Instruction Following Too close to call
GPT-5.4 mini: 74.3 (#99), Kimi K2 Thinking Turbo: 74.2 (#103)
| Benchmark | GPT-5.4 mini | Kimi K2 Thinking Turbo |
|---|---|---|
| LMArena Instruction Following | 1405 | 1403 |
Long Context Too close to call
GPT-5.4 mini: 42.9 (#107), Kimi K2 Thinking Turbo: 43.2 (#99)
| Benchmark | GPT-5.4 mini | Kimi K2 Thinking Turbo |
|---|---|---|
| LMArena Longer Query | 1407 | 1415 |
Writing & Preference GPT-5.4 mini leads
GPT-5.4 mini: 64.2 (#56), Kimi K2 Thinking Turbo: 60.2 (#99)
| Benchmark | GPT-5.4 mini | Kimi K2 Thinking Turbo |
|---|---|---|
| LMArena Text | 1411 | 1415 |
| LMArena Creative Writing | 1370 | 1373 |
| LMArena Multi-Turn | 1429 | 1414 |
| EQ-Bench Creative Writing | 1665 | — |
Frequently asked questions
Is GPT-5.4 mini better than Kimi K2 Thinking Turbo?
GPT-5.4 mini and Kimi K2 Thinking Turbo score almost the same on the Noometry Index (44.8 vs 45.4), so choose on price, context window or the category you care about most.
Is GPT-5.4 mini or Kimi K2 Thinking Turbo better for coding?
GPT-5.4 mini scores higher on coding benchmarks: 45.0 versus 38.1 in the Noometry coding category.
How many benchmarks do GPT-5.4 mini and Kimi K2 Thinking Turbo share?
21 benchmarks have published results for both models. GPT-5.4 mini has 46 scored results on Noometry and Kimi K2 Thinking Turbo has 21.