Model comparison
Mistral Small vs Phi-4 Mini
Mistral Small is the stronger model overall, scoring 33.1 to 30.4 on the Noometry Index. Phi-4 Mini costs 2.0× less per token, which makes it the better buy when Mistral Small's lead doesn't matter for your workload.
Last verified . 3 shared benchmarks.
Mistral AI
MicrosoftSummary
- They share 3 benchmarks with published results for both. Mistral Small scores higher in 2 categories and Phi-4 Mini in 1 category; 3 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Mistral Small leads 30.9 to 24.8.
- The biggest single-benchmark swing is Vectara Hallucination Rate: 5.1% for Mistral Small and 23.5% for Phi-4 Mini.
- Phi-4 Mini is cheaper at $0.075 / $0.30 per million input/output tokens, against $0.15 / $0.60 for Mistral Small.
- Mistral Small accepts more context: 256K tokens versus 128K.
Side by side
| Mistral Small | Phi-4 Mini | |
|---|---|---|
| Provider | Mistral AI | Microsoft |
| Noometry Index | 33.1 | 30.4 |
| Released | 2024-02-26 | 2024-12-11 |
| Weights | Open | Open |
| Context window | 256K | 128K |
| Max output | 256K | 4K |
| Input $ / M tokens | $0.15 | $0.075 |
| Output $ / M tokens | $0.60 | $0.30 |
| Results tracked | 39 | 3 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Mistral Small leads
Mistral Small: 33.5 (#242), Phi-4 Mini: 27.6 (#312)
| Benchmark | Mistral Small | Phi-4 Mini |
|---|---|---|
| SciCode | 26.5% | 10.8% |
| BigCodeBench Instruct | 36.1% | — |
| LiveBench Coding | 36.2% | — |
| LMArena Coding | 1363 | — |
| BigCodeBench Complete | 46.6% | — |
| ALE-Bench | 497.62 | — |
Agentic & Tool Use Not comparable
Mistral Small: 27.9 (#87), Phi-4 Mini: —
| Benchmark | Mistral Small | Phi-4 Mini |
|---|---|---|
| Berkeley Function Calling Leaderboard | 37.1% | — |
Reasoning Phi-4 Mini leads
Mistral Small: 19.3 (#238), Phi-4 Mini: 21.7 (#190)
| Benchmark | Mistral Small | Phi-4 Mini |
|---|---|---|
| CritPt | 0% | 0% |
| Kagi LLM Benchmark | 37.8% | — |
| LiveBench Reasoning | 44.8% | — |
| LMArena Hard Prompts | 1336 | — |
| DTBench | 70.9% | — |
| LiveBench Data Analysis | 53.7% | — |
| LMCA | 20.6% | — |
| LiveBench | 44% | — |
Math Not comparable
Mistral Small: 15.4 (#287), Phi-4 Mini: —
| Benchmark | Mistral Small | Phi-4 Mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 5.8% | — |
| LiveBench Math | 39.9% | — |
| LMArena Math | 1341 | — |
| MATH Level 5 | 46.8% | — |
Knowledge Mistral Small leads
Mistral Small: 30.9 (#214), Phi-4 Mini: 24.8 (#261)
| Benchmark | Mistral Small | Phi-4 Mini |
|---|---|---|
| Vectara Hallucination Rate | 5.1% | 23.5% |
| GPQA Diamond | 47.5% | — |
| LMArena Expert | 1293 | — |
| MMLU | 68.7% | — |
Multimodal Not comparable
Mistral Small: 32.9 (#91), Phi-4 Mini: —
| Benchmark | Mistral Small | Phi-4 Mini |
|---|---|---|
| LMArena Vision | 1142 | — |
Multilingual Not comparable
Mistral Small: 45.7 (#164), Phi-4 Mini: —
| Benchmark | Mistral Small | Phi-4 Mini |
|---|---|---|
| LMArena Non-English | 1315 | — |
| LMArena Chinese | 1342 | — |
| LMArena French | 1337 | — |
| LMArena German | 1341 | — |
| LMArena Japanese | 1275 | — |
| LMArena Korean | 1260 | — |
| LMArena Russian | 1324 | — |
| LMArena Spanish | 1346 | — |
Instruction Following Not comparable
Mistral Small: 66.6 (#204), Phi-4 Mini: —
| Benchmark | Mistral Small | Phi-4 Mini |
|---|---|---|
| LiveBench Instruction Following | 63.7% | — |
| LMArena Instruction Following | 1310 | — |
Long Context Not comparable
Mistral Small: 40.4 (#151), Phi-4 Mini: —
| Benchmark | Mistral Small | Phi-4 Mini |
|---|---|---|
| LMArena Longer Query | 1327 | — |
Writing & Preference Not comparable
Mistral Small: 52.6 (#165), Phi-4 Mini: —
| Benchmark | Mistral Small | Phi-4 Mini |
|---|---|---|
| LMArena Text | 1338 | — |
| LMArena Creative Writing | 1304 | — |
| LMArena Multi-Turn | 1344 | — |
| LiveBench Language | 30.5% | — |
Frequently asked questions
Is Mistral Small better than Phi-4 Mini?
Mistral Small is the stronger model overall, scoring 33.1 to 30.4 on the Noometry Index. Phi-4 Mini costs 2.0× less per token, which makes it the better buy when Mistral Small's lead doesn't matter for your workload.
Which is cheaper, Mistral Small or Phi-4 Mini?
Phi-4 Mini is cheaper. It lists at $0.075 per million input tokens and $0.30 per million output tokens; Mistral Small lists at $0.15 and $0.60.
Is Mistral Small or Phi-4 Mini better for coding?
Mistral Small scores higher on coding benchmarks: 33.5 versus 27.6 in the Noometry coding category.
Which has the bigger context window?
Mistral Small does, with 256K tokens against 128K.
How many benchmarks do Mistral Small and Phi-4 Mini share?
3 benchmarks have published results for both models. Mistral Small has 39 scored results on Noometry and Phi-4 Mini has 3.