Model comparison
Olmo 3 32b Think vs Phi-4
Olmo 3 32b Think is the stronger model overall, scoring 38.7 to 31.2 on the Noometry Index.
Last verified . 14 shared benchmarks.
Allen Institute for AI (Ai2)
MicrosoftSummary
- They share 14 benchmarks with published results for both. Olmo 3 32b Think scores higher in 8 categories and Phi-4 in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Olmo 3 32b Think leads 36.5 to 20.8.
Side by side
| Olmo 3 32b Think | Phi-4 | |
|---|---|---|
| Provider | Allen Institute for AI (Ai2) | Microsoft |
| Noometry Index | 38.7 | 31.2 |
| Released | — | 2024-12-11 |
| Weights | Open | Open |
| Context window | — | 128K |
| Max output | — | 4K |
| Input $ / M tokens | — | $0.07 |
| Output $ / M tokens | — | $0.14 |
| Results tracked | 14 | 37 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Olmo 3 32b Think leads
Olmo 3 32b Think: 38.6 (#172), Phi-4: 34.4 (#239)
| Benchmark | Olmo 3 32b Think | Phi-4 |
|---|---|---|
| LMArena Coding | 1319 | 1231 |
| BigCodeBench Instruct | — | 45.5% |
| LiveBench Coding | — | 30.7% |
| BigCodeBench Complete | — | 55.4% |
Agentic & Tool Use Not comparable
Olmo 3 32b Think: —, Phi-4: 22.8 (#128)
| Benchmark | Olmo 3 32b Think | Phi-4 |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 28.8% |
| BALROG | — | 11.6% |
Reasoning Olmo 3 32b Think leads
Olmo 3 32b Think: 25.9 (#140), Phi-4: 17.7 (#291)
| Benchmark | Olmo 3 32b Think | Phi-4 |
|---|---|---|
| LMArena Hard Prompts | 1302 | 1220 |
| Chess Puzzles | — | 1% |
| LiveBench Reasoning | — | 47.8% |
| LiveBench Data Analysis | — | 45.2% |
| Epoch Capabilities Index | — | 130.42 |
| LiveBench | — | 41.6% |
Math Olmo 3 32b Think leads
Olmo 3 32b Think: 36.5 (#165), Phi-4: 20.8 (#285)
| Benchmark | Olmo 3 32b Think | Phi-4 |
|---|---|---|
| LMArena Math | 1316 | 1246 |
| OTIS Mock AIME 2024-2025 | — | 13.8% |
| LiveBench Math | — | 42% |
| MATH Level 5 | — | 64.9% |
Knowledge Olmo 3 32b Think leads
Olmo 3 32b Think: 35.0 (#190), Phi-4: 32.6 (#209)
| Benchmark | Olmo 3 32b Think | Phi-4 |
|---|---|---|
| LMArena Expert | 1273 | 1203 |
| GPQA Diamond | — | 56.1% |
| Confabulations | — | 29.4% |
| Vectara Hallucination Rate | — | 3.7% |
| MMLU | — | 84.8% |
Multilingual Olmo 3 32b Think leads
Olmo 3 32b Think: 41.2 (#210), Phi-4: 37.2 (#237)
| Benchmark | Olmo 3 32b Think | Phi-4 |
|---|---|---|
| LMArena Non-English | 1255 | 1197 |
| LMArena Chinese | 1300 | 1212 |
| LMArena French | 1291 | 1224 |
| LMArena German | 1290 | 1222 |
| LMArena Russian | 1254 | 1209 |
| LMArena Japanese | — | 1158 |
| LMArena Korean | — | 1151 |
| LMArena Spanish | — | 1234 |
Instruction Following Olmo 3 32b Think leads
Olmo 3 32b Think: 67.2 (#198), Phi-4: 60.4 (#251)
| Benchmark | Olmo 3 32b Think | Phi-4 |
|---|---|---|
| LMArena Instruction Following | 1275 | 1201 |
| LiveBench Instruction Following | — | 58.4% |
Long Context Olmo 3 32b Think leads
Olmo 3 32b Think: 39.4 (#182), Phi-4: 36.9 (#226)
| Benchmark | Olmo 3 32b Think | Phi-4 |
|---|---|---|
| LMArena Longer Query | 1296 | 1217 |
Writing & Preference Olmo 3 32b Think leads
Olmo 3 32b Think: 49.1 (#193), Phi-4: 40.5 (#244)
| Benchmark | Olmo 3 32b Think | Phi-4 |
|---|---|---|
| LMArena Text | 1300 | 1217 |
| LMArena Creative Writing | 1256 | 1182 |
| LMArena Multi-Turn | 1290 | 1206 |
| Short-Story Creative Writing | — | 62.6% |
| LiveBench Language | — | 25.6% |
Frequently asked questions
Is Olmo 3 32b Think better than Phi-4?
Olmo 3 32b Think is the stronger model overall, scoring 38.7 to 31.2 on the Noometry Index.
Is Olmo 3 32b Think or Phi-4 better for coding?
Olmo 3 32b Think scores higher on coding benchmarks: 38.6 versus 34.4 in the Noometry coding category.
How many benchmarks do Olmo 3 32b Think and Phi-4 share?
14 benchmarks have published results for both models. Olmo 3 32b Think has 14 scored results on Noometry and Phi-4 has 37.