Model comparison
o3 vs Phi 3 Small 8k Instruct
o3 is the stronger model overall, scoring 47.5 to 29.3 on the Noometry Index.
Last verified . 17 shared benchmarks.
MicrosoftSummary
- They share 17 benchmarks with published results for both. o3 scores higher in 8 categories and Phi 3 Small 8k Instruct in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where o3 leads 63.5 to 31.1.
- Phi 3 Small 8k Instruct has downloadable open weights; the other is API-only.
Side by side
| o3 | Phi 3 Small 8k Instruct | |
|---|---|---|
| Provider | OpenAI | Microsoft |
| Noometry Index | 47.5 | 29.3 |
| Released | 2025-04-16 | 2024-04-23 |
| Weights | Proprietary | Open |
| Context window | 200K | — |
| Max output | 100K | — |
| Input $ / M tokens | $2 | — |
| Output $ / M tokens | $8 | — |
| Results tracked | 63 | 32 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding o3 leads
o3: 46.8 (#64), Phi 3 Small 8k Instruct: 27.9 (#318)
| Benchmark | o3 | Phi 3 Small 8k Instruct |
|---|---|---|
| LMArena Coding | 1408 | 1101 |
| SWE-bench Verified | 62.3% | — |
| SWE-bench Verified (bash only) | 58.4% | — |
| Aider Polyglot | 81.3% | — |
| GSO | 8.8% | — |
| WeirdML | 52.4% | — |
| LiveBench Coding | — | 20.3% |
| CadEval | 74% | — |
| ALE-Bench | 933.55 | — |
Agentic & Tool Use Not comparable
o3: 34.5 (#44), Phi 3 Small 8k Instruct: —
| Benchmark | o3 | Phi 3 Small 8k Instruct |
|---|---|---|
| Berkeley Function Calling Leaderboard | 63% | — |
| GDPval | 30.8% | — |
| DeepResearch Bench | 45.2% | — |
| OSWorld | 23% | — |
| LMArena Search | 1144 | — |
| METR Time Horizons | 65.4% | — |
Reasoning o3 leads
o3: 32.0 (#78), Phi 3 Small 8k Instruct: 14.8 (#323)
| Benchmark | o3 | Phi 3 Small 8k Instruct |
|---|---|---|
| LMArena Hard Prompts | 1402 | 1100 |
| ARC-AGI-2 | 6.5% | — |
| SimpleBench | 53.1% | — |
| Kagi LLM Benchmark | 67.6% | — |
| ARC-AGI-1 | 60.8% | — |
| CritPt | 1.4% | — |
| Chess Puzzles | 38% | — |
| EnigmaEval | 13.1% | — |
| LiveBench Reasoning | — | 15.9% |
| Mystery Game Puzzles | 29% | — |
| DTBench | 84.8% | — |
| LiveBench Data Analysis | — | 30.3% |
| LMCA | 39.7% | — |
| Adversarial NLI | — | 58.1% |
| BIG-Bench Hard | — | 79.1% |
| Epoch Capabilities Index | 146.86 | — |
| ForecastBench | 62.5 | — |
| HellaSwag | — | 77% |
| LiveBench | — | 24% |
| WinoGrande | — | 81.5% |
Math o3 leads
o3: 50.2 (#58), Phi 3 Small 8k Instruct: 27.6 (#248)
| Benchmark | o3 | Phi 3 Small 8k Instruct |
|---|---|---|
| LMArena Math | 1426 | 1151 |
| FrontierMath (Tiers 1-3) | 33.3% | — |
| OTIS Mock AIME 2024-2025 | 84.4% | — |
| Omni-MATH | 71.4% | — |
| LiveBench Math | — | 17.6% |
| MATH Level 5 | 97.8% | — |
| FrontierMath (Feb 2025 set) | 18.7% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge o3 leads
o3: 54.6 (#52), Phi 3 Small 8k Instruct: 29.1 (#240)
| Benchmark | o3 | Phi 3 Small 8k Instruct |
|---|---|---|
| LMArena Expert | 1402 | 1067 |
| GPQA Diamond | 81.8% | — |
| Humanity's Last Exam | 20.3% | — |
| SimpleQA Verified | 49.4% | — |
| MMLU-Pro | 85.9% | — |
| Confabulations | 14.4% | — |
| GPQA (HELM) | 75.3% | — |
| ARC (AI2) Challenge | — | 90.7% |
| MMLU | — | 75.7% |
| OpenBookQA | — | 88% |
| TriviaQA | — | 58.1% |
Multimodal Not comparable
o3: 41.4 (#36), Phi 3 Small 8k Instruct: —
| Benchmark | o3 | Phi 3 Small 8k Instruct |
|---|---|---|
| LMArena Vision | 1214 | — |
| GeoBench | 74% | — |
| VPCT | 52% | — |
Multilingual o3 leads
o3: 51.7 (#105), Phi 3 Small 8k Instruct: 28.5 (#272)
| Benchmark | o3 | Phi 3 Small 8k Instruct |
|---|---|---|
| LMArena Non-English | 1401 | 1058 |
| LMArena Chinese | 1437 | 1061 |
| LMArena French | 1430 | 1135 |
| LMArena German | 1420 | 1080 |
| LMArena Japanese | 1403 | 966 |
| LMArena Korean | 1370 | 894 |
| LMArena Russian | 1406 | 1111 |
| LMArena Spanish | 1395 | 1111 |
Instruction Following o3 leads
o3: 72.8 (#127), Phi 3 Small 8k Instruct: 51.9 (#292)
| Benchmark | o3 | Phi 3 Small 8k Instruct |
|---|---|---|
| LMArena Instruction Following | 1368 | 1087 |
| LiveBench Instruction Following | — | 47.2% |
| IFEval | 86.9% | — |
Long Context o3 leads
o3: 53.3 (#6), Phi 3 Small 8k Instruct: 33.0 (#267)
| Benchmark | o3 | Phi 3 Small 8k Instruct |
|---|---|---|
| LMArena Longer Query | 1372 | 1088 |
| Fiction.LiveBench | 88.9% | — |
| CL-bench | 17.8% | — |
Writing & Preference o3 leads
o3: 63.5 (#64), Phi 3 Small 8k Instruct: 31.1 (#284)
| Benchmark | o3 | Phi 3 Small 8k Instruct |
|---|---|---|
| LMArena Text | 1410 | 1110 |
| LMArena Creative Writing | 1359 | 1083 |
| LMArena Multi-Turn | 1405 | 1068 |
| Short-Story Creative Writing | 83.9% | — |
| EQ-Bench Creative Writing | 1676 | — |
| WildBench | 86.1% | — |
| LiveBench Language | — | 12.9% |
Frequently asked questions
Is o3 better than Phi 3 Small 8k Instruct?
o3 is the stronger model overall, scoring 47.5 to 29.3 on the Noometry Index.
Is o3 or Phi 3 Small 8k Instruct better for coding?
o3 scores higher on coding benchmarks: 46.8 versus 27.9 in the Noometry coding category.
How many benchmarks do o3 and Phi 3 Small 8k Instruct share?
17 benchmarks have published results for both models. o3 has 63 scored results on Noometry and Phi 3 Small 8k Instruct has 32.