Model comparison
GPT-4.1 mini vs Kimi K3
Kimi K3 is the stronger model overall, scoring 59.5 to 33.5 on the Noometry Index. GPT-4.1 mini costs 8.6× less per token, which makes it the better buy when Kimi K3's lead doesn't matter for your workload.
Last verified . 32 shared benchmarks.
Summary
- They share 32 benchmarks with published results for both. GPT-4.1 mini scores higher in 0 categories and Kimi K3 in 10 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Kimi K3 leads 63.0 to 10.9.
- The biggest single-benchmark swing is ARC-AGI-1: 3.5% for GPT-4.1 mini and 94.5% for Kimi K3.
- GPT-4.1 mini is cheaper at $0.40 / $1.60 per million input/output tokens, against $3 / $15 for Kimi K3.
- Kimi K3 accepts more context: 1.05M tokens versus 1.05M.
- Kimi K3 has downloadable open weights; the other is API-only.
Side by side
| GPT-4.1 mini | Kimi K3 | |
|---|---|---|
| Provider | OpenAI | Moonshot AI |
| Noometry Index | 33.5 | 59.5 |
| Released | 2025-04-14 | 2026-07-16 |
| Weights | Proprietary | Open |
| Context window | 1.05M | 1.05M |
| Max output | 33K | 1.05M |
| Input $ / M tokens | $0.40 | $3 |
| Output $ / M tokens | $1.60 | $15 |
| Results tracked | 47 | 53 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K3 leads
GPT-4.1 mini: 30.4 (#286), Kimi K3: 60.6 (#8)
| Benchmark | GPT-4.1 mini | Kimi K3 |
|---|---|---|
| SciCode | 40.4% | 59.5% |
| WeirdML | 37.6% | 82.6% |
| LMArena Coding | 1367 | 1509 |
| DeepSWE | — | 68.5% |
| FrontierCode | — | 44.2% |
| SWE-bench Verified (bash only) | 23.9% | — |
| Aider Polyglot | 32.4% | — |
| LMArena WebDev | — | 1659 |
| FrontierSWE | — | 25.9% |
| BigCodeBench Instruct | 48.9% | — |
| CadEval | 16% | — |
| ALE-Bench | — | 1,524 |
Agentic & Tool Use Kimi K3 leads
GPT-4.1 mini: 33.0 (#52), Kimi K3: 41.8 (#19)
| Benchmark | GPT-4.1 mini | Kimi K3 |
|---|---|---|
| APEX-Agents | — | 50.6% |
| Berkeley Function Calling Leaderboard | 50.5% | — |
| τ²-bench Banking | — | 37.1% |
| PostTrainBench | — | 32% |
| GBAEval | — | 48.3% |
| GDP.pdf | — | 19% |
| Vending-Bench 2 | — | 5,165 |
Reasoning Kimi K3 leads
GPT-4.1 mini: 10.9 (#335), Kimi K3: 63.0 (#15)
| Benchmark | GPT-4.1 mini | Kimi K3 |
|---|---|---|
| ARC-AGI-2 | 0% | 60.4% |
| ARC-AGI-1 | 3.5% | 94.5% |
| CritPt | 0% | 23.4% |
| Chess Puzzles | 7% | 39% |
| LMArena Hard Prompts | 1349 | 1495 |
| Mystery Game Puzzles | 7% | 26% |
| DTBench | 68.8% | 91.2% |
| LMCA | 21.1% | 52.7% |
| Epoch Capabilities Index | 135.01 | 157.68 |
| SimpleBench | — | 60.7% |
| Kagi LLM Benchmark | 48.6% | — |
| NYT Connections (extended) | — | 93.6% |
| Surface Evolver Bench | — | 95% |
| ForecastBench | — | 61.1 |
Math Kimi K3 leads
GPT-4.1 mini: 23.1 (#264), Kimi K3: 74.0 (#13)
| Benchmark | GPT-4.1 mini | Kimi K3 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 6.7% | 72.2% |
| OTIS Mock AIME 2024-2025 | 44.7% | 97.2% |
| LMArena Math | 1343 | 1493 |
| FrontierMath Tier 4 | — | 39% |
| MathArena Final-Answer Competitions | — | 87.8% |
| ProofBench | — | 87% |
| Omni-MATH | 49.1% | — |
| MATH Level 5 | 87.3% | — |
| FrontierMath (Feb 2025 set) | 4.5% | — |
Knowledge Kimi K3 leads
GPT-4.1 mini: 34.7 (#183), Kimi K3: 63.2 (#19)
| Benchmark | GPT-4.1 mini | Kimi K3 |
|---|---|---|
| GPQA Diamond | 65.8% | 93.1% |
| SimpleQA Verified | 12.7% | 50.6% |
| LMArena Expert | 1338 | 1520 |
| MMLU-Pro | 78.3% | — |
| GPQA (HELM) | 61.4% | — |
Multimodal Kimi K3 leads
GPT-4.1 mini: 35.2 (#77), Kimi K3: 38.4 (#58)
| Benchmark | GPT-4.1 mini | Kimi K3 |
|---|---|---|
| LMArena Vision | 1181 | — |
| Blueprint-Bench 2 | — | 29.5% |
| Furniture Assembly | — | 34.2% |
Multilingual Kimi K3 leads
GPT-4.1 mini: 45.9 (#161), Kimi K3: 56.5 (#20)
| Benchmark | GPT-4.1 mini | Kimi K3 |
|---|---|---|
| LMArena Non-English | 1318 | 1466 |
| LMArena Chinese | 1329 | 1528 |
| LMArena French | 1357 | 1488 |
| LMArena German | 1349 | 1474 |
| LMArena Japanese | 1291 | 1492 |
| LMArena Korean | 1298 | 1465 |
| LMArena Russian | 1324 | 1481 |
| LMArena Spanish | 1319 | 1478 |
Instruction Following Kimi K3 leads
GPT-4.1 mini: 73.9 (#113), Kimi K3: 77.8 (#13)
| Benchmark | GPT-4.1 mini | Kimi K3 |
|---|---|---|
| LMArena Instruction Following | 1332 | 1481 |
| IFEval | 90.4% | — |
Long Context Kimi K3 leads
GPT-4.1 mini: 31.8 (#270), Kimi K3: 45.8 (#27)
| Benchmark | GPT-4.1 mini | Kimi K3 |
|---|---|---|
| LMArena Longer Query | 1344 | 1494 |
| Fiction.LiveBench | 44.4% | — |
Writing & Preference Kimi K3 leads
GPT-4.1 mini: 48.8 (#193), Kimi K3: 76.7 (#4)
| Benchmark | GPT-4.1 mini | Kimi K3 |
|---|---|---|
| LMArena Text | 1341 | 1476 |
| LMArena Creative Writing | 1300 | 1452 |
| EQ-Bench Creative Writing | 1147 | 2082 |
| LMArena Multi-Turn | 1354 | 1488 |
| WildBench | 83.8% | — |
| EQ-Bench 4 | — | 1339 |
Frequently asked questions
Is GPT-4.1 mini better than Kimi K3?
Kimi K3 is the stronger model overall, scoring 59.5 to 33.5 on the Noometry Index. GPT-4.1 mini costs 8.6× less per token, which makes it the better buy when Kimi K3's lead doesn't matter for your workload.
Which is cheaper, GPT-4.1 mini or Kimi K3?
GPT-4.1 mini is cheaper. It lists at $0.40 per million input tokens and $1.60 per million output tokens; Kimi K3 lists at $3 and $15.
Is GPT-4.1 mini or Kimi K3 better for coding?
Kimi K3 scores higher on coding benchmarks: 60.6 versus 30.4 in the Noometry coding category.
Which has the bigger context window?
Kimi K3 does, with 1.05M tokens against 1.05M.
How many benchmarks do GPT-4.1 mini and Kimi K3 share?
32 benchmarks have published results for both models. GPT-4.1 mini has 47 scored results on Noometry and Kimi K3 has 53.