Model comparison
Gemini 3 Flash Preview vs Phi 3 Mini 4k Instruct
Gemini 3 Flash Preview is the stronger model overall, scoring 52.3 to 27.9 on the Noometry Index.
Last verified . 18 shared benchmarks.
Google
MicrosoftSummary
- They share 18 benchmarks with published results for both. Gemini 3 Flash Preview scores higher in 8 categories and Phi 3 Mini 4k Instruct in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Gemini 3 Flash Preview leads 65.5 to 27.6.
- The biggest single-benchmark swing is Chess Puzzles: 40% for Gemini 3 Flash Preview and 0% for Phi 3 Mini 4k Instruct.
- Phi 3 Mini 4k Instruct has downloadable open weights; the other is API-only.
Side by side
| Gemini 3 Flash Preview | Phi 3 Mini 4k Instruct | |
|---|---|---|
| Provider | Microsoft | |
| Noometry Index | 52.3 | 27.9 |
| Released | 2025-12-17 | 2024-04-23 |
| Weights | Proprietary | Open |
| Context window | 1.05M | — |
| Max output | 66K | — |
| Input $ / M tokens | $0.50 | — |
| Output $ / M tokens | $3 | — |
| Results tracked | 59 | 35 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemini 3 Flash Preview leads
Gemini 3 Flash Preview: 50.9 (#42), Phi 3 Mini 4k Instruct: 26.6 (#323)
| Benchmark | Gemini 3 Flash Preview | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Coding | 1460 | 1093 |
| SWE-bench Verified | 75.4% | — |
| SWE-bench Verified (bash only) | 75.8% | — |
| LMArena WebDev | 1439 | — |
| SWE-bench Multilingual | 72.7% | — |
| GSO | 9.8% | — |
| WeirdML | 61.6% | — |
| LiveBench Coding | — | 15.5% |
| ALE-Bench | 1,367 | — |
| HumanEval+ | — | 59.1% |
| MBPP+ | — | 54.2% |
Agentic & Tool Use Not comparable
Gemini 3 Flash Preview: 38.7 (#29), Phi 3 Mini 4k Instruct: —
| Benchmark | Gemini 3 Flash Preview | Phi 3 Mini 4k Instruct |
|---|---|---|
| Terminal-Bench | 64.3% | — |
| τ²-bench Airline | 82.5% | — |
| τ²-bench Banking | 27.3% | — |
| τ²-bench Retail | 76.8% | — |
| τ²-bench Telecom | 91.2% | — |
| DeepResearch Bench | 49.8% | — |
| BALROG | 48.1% | — |
| GDP.pdf | 10% | — |
| LMArena Search | 1198 | — |
| Vending-Bench 2 | 3,635 | — |
Reasoning Gemini 3 Flash Preview leads
Gemini 3 Flash Preview: 49.2 (#37), Phi 3 Mini 4k Instruct: 14.1 (#328)
| Benchmark | Gemini 3 Flash Preview | Phi 3 Mini 4k Instruct |
|---|---|---|
| Chess Puzzles | 40% | 0% |
| LMArena Hard Prompts | 1465 | 1072 |
| ARC-AGI-2 | 33.6% | — |
| SimpleBench | 61.1% | — |
| NYT Connections (extended) | 83.1% | — |
| ARC-AGI-1 | 84.7% | — |
| LiveBench Reasoning | — | 26.8% |
| Mystery Game Puzzles | 26% | — |
| DTBench | 89.1% | — |
| LiveBench Data Analysis | — | 34.7% |
| LMCA | 43.1% | — |
| Adversarial NLI | — | 52.8% |
| BIG-Bench Hard | — | 71.7% |
| Epoch Capabilities Index | 151.8 | — |
| ForecastBench | 58.5 | — |
| HellaSwag | — | 76.7% |
| LiveBench | — | 22.4% |
| WinoGrande | — | 70.8% |
Math Gemini 3 Flash Preview leads
Gemini 3 Flash Preview: 51.7 (#55), Phi 3 Mini 4k Instruct: 26.6 (#257)
| Benchmark | Gemini 3 Flash Preview | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Math | 1473 | 1111 |
| FrontierMath (Tiers 1-3) | 51.2% | — |
| FrontierMath Tier 4 | 17.1% | — |
| MathArena Final-Answer Competitions | 67.6% | — |
| OTIS Mock AIME 2024-2025 | 95.6% | — |
| ProofBench | 15% | — |
| LiveBench Math | — | 15.7% |
| FrontierMath (Feb 2025 set) | 35.6% | — |
| FrontierMath Tier 4 (v1) | 4.2% | — |
Knowledge Gemini 3 Flash Preview leads
Gemini 3 Flash Preview: 58.8 (#33), Phi 3 Mini 4k Instruct: 28.5 (#246)
| Benchmark | Gemini 3 Flash Preview | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Expert | 1462 | 1045 |
| GPQA Diamond | 89.4% | — |
| SimpleQA Verified | 66.8% | — |
| Vectara Hallucination Rate | 13.5% | — |
| ARC (AI2) Challenge | — | 84.9% |
| MMLU | — | 68.8% |
| OpenBookQA | — | 88% |
| TriviaQA | — | 64% |
Multimodal Not comparable
Gemini 3 Flash Preview: 45.5 (#16), Phi 3 Mini 4k Instruct: —
| Benchmark | Gemini 3 Flash Preview | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Vision | 1285 | — |
| GeoBench | 88% | — |
| VPCT | 72.6% | — |
| Blueprint-Bench 2 | 0% | — |
| LMArena Document | 1413 | — |
Multilingual Gemini 3 Flash Preview leads
Gemini 3 Flash Preview: 55.7 (#27), Phi 3 Mini 4k Instruct: 26.3 (#280)
| Benchmark | Gemini 3 Flash Preview | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Non-English | 1458 | 1021 |
| LMArena Chinese | 1511 | 1021 |
| LMArena French | 1477 | 1076 |
| LMArena German | 1497 | 1044 |
| LMArena Japanese | 1489 | 935 |
| LMArena Korean | 1443 | 905 |
| LMArena Russian | 1480 | 1022 |
| LMArena Spanish | 1469 | 1085 |
Instruction Following Gemini 3 Flash Preview leads
Gemini 3 Flash Preview: 75.7 (#56), Phi 3 Mini 4k Instruct: 47.7 (#303)
| Benchmark | Gemini 3 Flash Preview | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Instruction Following | 1437 | 1053 |
| LiveBench Instruction Following | — | 39.1% |
Long Context Gemini 3 Flash Preview leads
Gemini 3 Flash Preview: 44.4 (#67), Phi 3 Mini 4k Instruct: 31.7 (#276)
| Benchmark | Gemini 3 Flash Preview | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Longer Query | 1452 | 1044 |
Writing & Preference Gemini 3 Flash Preview leads
Gemini 3 Flash Preview: 65.5 (#45), Phi 3 Mini 4k Instruct: 27.6 (#300)
| Benchmark | Gemini 3 Flash Preview | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Text | 1466 | 1073 |
| LMArena Creative Writing | 1457 | 1037 |
| LMArena Multi-Turn | 1471 | 1018 |
| LiveBench Language | — | 9.2% |
Frequently asked questions
Is Gemini 3 Flash Preview better than Phi 3 Mini 4k Instruct?
Gemini 3 Flash Preview is the stronger model overall, scoring 52.3 to 27.9 on the Noometry Index.
Is Gemini 3 Flash Preview or Phi 3 Mini 4k Instruct better for coding?
Gemini 3 Flash Preview scores higher on coding benchmarks: 50.9 versus 26.6 in the Noometry coding category.
How many benchmarks do Gemini 3 Flash Preview and Phi 3 Mini 4k Instruct share?
18 benchmarks have published results for both models. Gemini 3 Flash Preview has 59 scored results on Noometry and Phi 3 Mini 4k Instruct has 35.