Model comparison
Mistral Small 3 vs Phi-4
Mistral Small 3 and Phi-4 score almost the same on the Noometry Index (31.2 vs 31.2), so choose on price, context window or the category you care about most.
Last verified . 23 shared benchmarks.
Mistral AI
MicrosoftSummary
- They share 23 benchmarks with published results for both. Mistral Small 3 scores higher in 5 categories and Phi-4 in 3 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Phi-4 leads 40.5 to 32.2.
- The biggest single-benchmark swing is GPQA Diamond: 47.3% for Mistral Small 3 and 56.1% for Phi-4.
- Mistral Small 3 is cheaper at $0.05 / $0.08 per million input/output tokens, against $0.07 / $0.14 for Phi-4.
- Phi-4 accepts more context: 128K tokens versus 33K.
Side by side
| Mistral Small 3 | Phi-4 | |
|---|---|---|
| Provider | Mistral AI | Microsoft |
| Noometry Index | 31.2 | 31.2 |
| Released | 2025-01-30 | 2024-12-11 |
| Weights | Open | Open |
| Context window | 33K | 128K |
| Max output | 16K | 4K |
| Input $ / M tokens | $0.05 | $0.07 |
| Output $ / M tokens | $0.08 | $0.14 |
| Results tracked | 24 | 37 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Mistral Small 3 leads
Mistral Small 3: 36.5 (#207), Phi-4: 34.4 (#239)
| Benchmark | Mistral Small 3 | Phi-4 |
|---|---|---|
| BigCodeBench Instruct | 45.3% | 45.5% |
| LMArena Coding | 1246 | 1231 |
| BigCodeBench Complete | 50.4% | 55.4% |
| LiveBench Coding | — | 30.7% |
Agentic & Tool Use Not comparable
Mistral Small 3: —, Phi-4: 22.8 (#128)
| Benchmark | Mistral Small 3 | Phi-4 |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 28.8% |
| BALROG | — | 11.6% |
Reasoning Mistral Small 3 leads
Mistral Small 3: 18.9 (#273), Phi-4: 17.7 (#291)
| Benchmark | Mistral Small 3 | Phi-4 |
|---|---|---|
| Chess Puzzles | 0% | 1% |
| LMArena Hard Prompts | 1233 | 1220 |
| Epoch Capabilities Index | 127.07 | 130.42 |
| LiveBench Reasoning | — | 47.8% |
| LiveBench Data Analysis | — | 45.2% |
| LiveBench | — | 41.6% |
Math Phi-4 leads
Mistral Small 3: 16.3 (#295), Phi-4: 20.8 (#285)
| Benchmark | Mistral Small 3 | Phi-4 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 6.7% | 13.8% |
| LMArena Math | 1240 | 1246 |
| LiveBench Math | — | 42% |
| MATH Level 5 | — | 64.9% |
Knowledge Phi-4 leads
Mistral Small 3: 25.1 (#263), Phi-4: 32.6 (#209)
| Benchmark | Mistral Small 3 | Phi-4 |
|---|---|---|
| GPQA Diamond | 47.3% | 56.1% |
| Confabulations | 25.2% | 29.4% |
| LMArena Expert | 1202 | 1203 |
| Vectara Hallucination Rate | — | 3.7% |
| MMLU | — | 84.8% |
Multilingual Too close to call
Mistral Small 3: 37.3 (#236), Phi-4: 37.2 (#237)
| Benchmark | Mistral Small 3 | Phi-4 |
|---|---|---|
| LMArena Non-English | 1198 | 1197 |
| LMArena Chinese | 1204 | 1212 |
| LMArena French | 1203 | 1224 |
| LMArena German | 1211 | 1222 |
| LMArena Japanese | 1111 | 1158 |
| LMArena Korean | 1188 | 1151 |
| LMArena Russian | 1216 | 1209 |
| LMArena Spanish | — | 1234 |
Instruction Following Mistral Small 3 leads
Mistral Small 3: 63.7 (#229), Phi-4: 60.4 (#251)
| Benchmark | Mistral Small 3 | Phi-4 |
|---|---|---|
| LMArena Instruction Following | 1214 | 1201 |
| LiveBench Instruction Following | — | 58.4% |
Long Context Too close to call
Mistral Small 3: 37.8 (#211), Phi-4: 36.9 (#226)
| Benchmark | Mistral Small 3 | Phi-4 |
|---|---|---|
| LMArena Longer Query | 1246 | 1217 |
Writing & Preference Phi-4 leads
Mistral Small 3: 32.2 (#280), Phi-4: 40.5 (#244)
| Benchmark | Mistral Small 3 | Phi-4 |
|---|---|---|
| LMArena Text | 1234 | 1217 |
| LMArena Creative Writing | 1195 | 1182 |
| LMArena Multi-Turn | 1217 | 1206 |
| Short-Story Creative Writing | — | 62.6% |
| EQ-Bench Creative Writing | 707 | — |
| LiveBench Language | — | 25.6% |
Frequently asked questions
Is Mistral Small 3 better than Phi-4?
Mistral Small 3 and Phi-4 score almost the same on the Noometry Index (31.2 vs 31.2), so choose on price, context window or the category you care about most.
Which is cheaper, Mistral Small 3 or Phi-4?
Mistral Small 3 is cheaper. It lists at $0.05 per million input tokens and $0.08 per million output tokens; Phi-4 lists at $0.07 and $0.14.
Is Mistral Small 3 or Phi-4 better for coding?
Mistral Small 3 scores higher on coding benchmarks: 36.5 versus 34.4 in the Noometry coding category.
Which has the bigger context window?
Phi-4 does, with 128K tokens against 33K.
How many benchmarks do Mistral Small 3 and Phi-4 share?
23 benchmarks have published results for both models. Mistral Small 3 has 24 scored results on Noometry and Phi-4 has 37.