Model comparison
Gemini 2.5 Flash vs Grok-3 mini
Grok-3 mini is the stronger model overall, scoring 41.2 to 39.3 on the Noometry Index.
Last verified . 33 shared benchmarks.
GoogleSummary
- They share 33 benchmarks with published results for both. Gemini 2.5 Flash scores higher in 4 categories and Grok-3 mini in 4 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Grok-3 mini leads 46.4 to 36.4.
- The biggest single-benchmark swing is GPQA (HELM): 39% for Gemini 2.5 Flash and 67.5% for Grok-3 mini.
Side by side
| Gemini 2.5 Flash | Grok-3 mini | |
|---|---|---|
| Provider | xAI | |
| Noometry Index | 39.3 | 41.2 |
| Released | 2025-04-17 | 2025-04-09 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | — |
| Max output | 66K | — |
| Input $ / M tokens | $0.30 | — |
| Output $ / M tokens | $2.50 | — |
| Results tracked | 54 | 35 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok-3 mini leads
Gemini 2.5 Flash: 35.8 (#220), Grok-3 mini: 40.8 (#131)
| Benchmark | Gemini 2.5 Flash | Grok-3 mini |
|---|---|---|
| Aider Polyglot | 55.1% | 49.3% |
| WeirdML | 41.9% | 42.6% |
| LMArena Coding | 1424 | 1379 |
| SWE-bench Verified (bash only) | 28.7% | — |
| ALE-Bench | 661.88 | — |
Agentic & Tool Use Not comparable
Gemini 2.5 Flash: 30.8 (#74), Grok-3 mini: —
| Benchmark | Gemini 2.5 Flash | Grok-3 mini |
|---|---|---|
| Terminal-Bench | 17.1% | — |
| Berkeley Function Calling Leaderboard | 56.2% | — |
| TheAgentCompany | 41.1% | — |
| BALROG | 33.5% | — |
| Vending-Bench 2 | 548.84 | — |
Reasoning Gemini 2.5 Flash leads
Gemini 2.5 Flash: 18.1 (#286), Grok-3 mini: 13.6 (#334)
| Benchmark | Gemini 2.5 Flash | Grok-3 mini |
|---|---|---|
| ARC-AGI-2 | 2.5% | 0.4% |
| Kagi LLM Benchmark | 56.8% | 61.3% |
| ARC-AGI-1 | 33.3% | 16.5% |
| LMArena Hard Prompts | 1422 | 1375 |
| Epoch Capabilities Index | 143.03 | 140.35 |
| SimpleBench | 41.2% | — |
| CritPt | 1.1% | — |
| EnigmaEval | 2.7% | — |
| DTBench | 76.5% | — |
| LMCA | 27.5% | — |
| ForecastBench | 60.6 | — |
Math Grok-3 mini leads
Gemini 2.5 Flash: 39.9 (#98), Grok-3 mini: 42.1 (#85)
| Benchmark | Gemini 2.5 Flash | Grok-3 mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 73.1% | 77.8% |
| Omni-MATH | 38.5% | 31.8% |
| LMArena Math | 1415 | 1386 |
| FrontierMath (Feb 2025 set) | 4.8% | 5.9% |
| MATH Level 5 | — | 90.9% |
| FrontierMath Tier 4 (v1) | 4.2% | — |
Knowledge Grok-3 mini leads
Gemini 2.5 Flash: 36.4 (#168), Grok-3 mini: 46.4 (#81)
| Benchmark | Gemini 2.5 Flash | Grok-3 mini |
|---|---|---|
| MMLU-Pro | 63.9% | 79.9% |
| Confabulations | 16.8% | 10.8% |
| GPQA (HELM) | 39% | 67.5% |
| LMArena Expert | 1426 | 1395 |
| GPQA Diamond | — | 76.3% |
| Humanity's Last Exam | 12.1% | — |
| Vectara Hallucination Rate | 7.8% | — |
Multimodal Not comparable
Gemini 2.5 Flash: 41.8 (#32), Grok-3 mini: —
| Benchmark | Gemini 2.5 Flash | Grok-3 mini |
|---|---|---|
| LMArena Vision | 1253 | — |
| GeoBench | 76% | — |
| VPCT | 46.2% | — |
| SpatialViz-Bench | 36.9% | — |
Multilingual Gemini 2.5 Flash leads
Gemini 2.5 Flash: 52.3 (#88), Grok-3 mini: 48.1 (#145)
| Benchmark | Gemini 2.5 Flash | Grok-3 mini |
|---|---|---|
| LMArena Non-English | 1409 | 1352 |
| LMArena Chinese | 1450 | 1387 |
| LMArena French | 1433 | 1357 |
| LMArena German | 1418 | 1349 |
| LMArena Japanese | 1405 | 1342 |
| LMArena Korean | 1385 | 1335 |
| LMArena Russian | 1415 | 1353 |
| LMArena Spanish | 1421 | 1381 |
Instruction Following Grok-3 mini leads
Gemini 2.5 Flash: 75.7 (#54), Grok-3 mini: 78.5 (#9)
| Benchmark | Gemini 2.5 Flash | Grok-3 mini |
|---|---|---|
| IFEval | 89.8% | 95.1% |
| LMArena Instruction Following | 1405 | 1357 |
Long Context Gemini 2.5 Flash leads
Gemini 2.5 Flash: 47.5 (#17), Grok-3 mini: 41.0 (#147)
| Benchmark | Gemini 2.5 Flash | Grok-3 mini |
|---|---|---|
| Fiction.LiveBench | 77.8% | 66.7% |
| LMArena Longer Query | 1419 | 1372 |
Writing & Preference Gemini 2.5 Flash leads
Gemini 2.5 Flash: 53.8 (#157), Grok-3 mini: 52.5 (#169)
| Benchmark | Gemini 2.5 Flash | Grok-3 mini |
|---|---|---|
| LMArena Text | 1417 | 1370 |
| LMArena Creative Writing | 1400 | 1342 |
| Short-Story Creative Writing | 76.5% | 73.5% |
| WildBench | 81.7% | 65.1% |
| LMArena Multi-Turn | 1408 | 1355 |
| EQ-Bench Creative Writing | 1137 | — |
Frequently asked questions
Is Gemini 2.5 Flash better than Grok-3 mini?
Grok-3 mini is the stronger model overall, scoring 41.2 to 39.3 on the Noometry Index.
Is Gemini 2.5 Flash or Grok-3 mini better for coding?
Grok-3 mini scores higher on coding benchmarks: 40.8 versus 35.8 in the Noometry coding category.
How many benchmarks do Gemini 2.5 Flash and Grok-3 mini share?
33 benchmarks have published results for both models. Gemini 2.5 Flash has 54 scored results on Noometry and Grok-3 mini has 35.