Model comparison
GPT-5 Mini vs Grok-3 mini
GPT-5 Mini and Grok-3 mini score almost the same on the Noometry Index (41.7 vs 41.0), so choose on price, context window or the category you care about most.
Last verified . 34 shared benchmarks.
Summary
- They share 34 benchmarks with published results for both. GPT-5 Mini scores higher in 5 categories and Grok-3 mini in 3 categories; 4 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GPT-5 Mini leads 24.1 to 13.8.
- The biggest single-benchmark swing is Omni-MATH: 72.2% for GPT-5 Mini and 31.8% for Grok-3 mini.
Side by side
| GPT-5 Mini | Grok-3 mini | |
|---|---|---|
| Provider | OpenAI | xAI |
| Noometry Index | 41.7 | 41.0 |
| Released | 2025-08-07 | 2025-04-09 |
| Weights | Proprietary | Proprietary |
| Context window | 400K | — |
| Max output | 128K | — |
| Input $ / M tokens | $0.25 | — |
| Output $ / M tokens | $2 | — |
| Results tracked | 60 | 35 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
GPT-5 Mini: 40.0 (#134), Grok-3 mini: 40.4 (#126)
| Benchmark | GPT-5 Mini | Grok-3 mini |
|---|---|---|
| WeirdML | 52.7% | 42.6% |
| LMArena Coding | 1406 | 1380 |
| SWE-bench Verified | 64.7% | — |
| SWE-bench Verified (bash only) | 59.8% | — |
| Aider Polyglot | — | 49.3% |
| SWE-bench Multilingual | 39.7% | — |
| SciCode | 39.2% | — |
| ALE-Bench | 799.77 | — |
| AlgoTune | 1.38 | — |
Agentic & Tool Use Not comparable
GPT-5 Mini: 31.0 (#64), Grok-3 mini: —
| Benchmark | GPT-5 Mini | Grok-3 mini |
|---|---|---|
| Terminal-Bench | 34.8% | — |
| Berkeley Function Calling Leaderboard | 55.5% | — |
| Vending-Bench 2 | -31.18 | — |
Reasoning GPT-5 Mini leads
GPT-5 Mini: 24.1 (#153), Grok-3 mini: 13.8 (#323)
| Benchmark | GPT-5 Mini | Grok-3 mini |
|---|---|---|
| ARC-AGI-2 | 4.4% | 0.4% |
| Kagi LLM Benchmark | 70.3% | 61.3% |
| ARC-AGI-1 | 54.3% | 16.5% |
| LMArena Hard Prompts | 1380 | 1374 |
| Epoch Capabilities Index | 145.51 | 140.35 |
| CritPt | 0% | — |
| Chess Puzzles | 30% | — |
| EnigmaEval | 8.2% | — |
| Mystery Game Puzzles | 10% | — |
| DTBench | 80.5% | — |
| LMCA | 34.2% | — |
| ForecastBench | 61 | — |
Math GPT-5 Mini leads
GPT-5 Mini: 46.3 (#59), Grok-3 mini: 40.3 (#78)
| Benchmark | GPT-5 Mini | Grok-3 mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 86.7% | 77.8% |
| Omni-MATH | 72.2% | 31.8% |
| LMArena Math | 1377 | 1386 |
| MATH Level 5 | 97.8% | 90.9% |
| FrontierMath (Feb 2025 set) | 27.2% | 5.9% |
| FrontierMath (Tiers 1-3) | 46.7% | — |
| FrontierMath Tier 4 | 12.2% | — |
| ProofBench | 9% | — |
| FrontierMath Tier 4 (v1) | 6.3% | — |
Knowledge Too close to call
GPT-5 Mini: 45.5 (#80), Grok-3 mini: 46.3 (#75)
| Benchmark | GPT-5 Mini | Grok-3 mini |
|---|---|---|
| GPQA Diamond | 75% | 76.3% |
| MMLU-Pro | 83.5% | 79.9% |
| Confabulations | 13.3% | 10.8% |
| GPQA (HELM) | 75.6% | 67.5% |
| LMArena Expert | 1377 | 1395 |
| Humanity's Last Exam | 19.4% | — |
| SimpleQA Verified | 21.6% | — |
| Vectara Hallucination Rate | 12.9% | — |
Multimodal Not comparable
GPT-5 Mini: 35.0 (#80), Grok-3 mini: —
| Benchmark | GPT-5 Mini | Grok-3 mini |
|---|---|---|
| LMArena Vision | 1202 | — |
| VPCT | 40.2% | — |
Multilingual Too close to call
GPT-5 Mini: 49.1 (#131), Grok-3 mini: 48.3 (#141)
| Benchmark | GPT-5 Mini | Grok-3 mini |
|---|---|---|
| LMArena Non-English | 1363 | 1352 |
| LMArena Chinese | 1387 | 1387 |
| LMArena French | 1385 | 1357 |
| LMArena German | 1366 | 1350 |
| LMArena Japanese | 1342 | 1343 |
| LMArena Korean | 1306 | 1336 |
| LMArena Russian | 1362 | 1352 |
| LMArena Spanish | 1355 | 1379 |
Instruction Following Grok-3 mini leads
GPT-5 Mini: 76.4 (#43), Grok-3 mini: 78.7 (#8)
| Benchmark | GPT-5 Mini | Grok-3 mini |
|---|---|---|
| IFEval | 92.7% | 95.1% |
| LMArena Instruction Following | 1357 | 1356 |
Long Context Too close to call
GPT-5 Mini: 41.9 (#127), Grok-3 mini: 41.0 (#142)
| Benchmark | GPT-5 Mini | Grok-3 mini |
|---|---|---|
| Fiction.LiveBench | 69.4% | 66.7% |
| LMArena Longer Query | 1355 | 1372 |
Writing & Preference GPT-5 Mini leads
GPT-5 Mini: 55.4 (#143), Grok-3 mini: 52.7 (#163)
| Benchmark | GPT-5 Mini | Grok-3 mini |
|---|---|---|
| LMArena Text | 1373 | 1370 |
| LMArena Creative Writing | 1325 | 1342 |
| Short-Story Creative Writing | 83.1% | 73.5% |
| WildBench | 85.5% | 65.1% |
| LMArena Multi-Turn | 1363 | 1355 |
| EQ-Bench Creative Writing | 1313 | — |
Frequently asked questions
Is GPT-5 Mini better than Grok-3 mini?
GPT-5 Mini and Grok-3 mini score almost the same on the Noometry Index (41.7 vs 41.0), so choose on price, context window or the category you care about most.
Is GPT-5 Mini or Grok-3 mini better for coding?
They score almost the same on coding (40.0 vs 40.4); test both on your own repository before choosing.
How many benchmarks do GPT-5 Mini and Grok-3 mini share?
34 benchmarks have published results for both models. GPT-5 Mini has 60 scored results on Noometry and Grok-3 mini has 35.