Model comparison
Ministral 8B vs phi-3-medium 14B
phi-3-medium 14B is the stronger model overall, scoring 29.7 to 28.2 on the Noometry Index.
Last verified . 2 shared benchmarks.
Mistral AI
MicrosoftSummary
- They share 2 benchmarks with published results for both. Ministral 8B scores higher in 1 category and phi-3-medium 14B in 2 categories; 3 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Ministral 8B leads 12.6 to 9.1.
Side by side
| Ministral 8B | phi-3-medium 14B | |
|---|---|---|
| Provider | Mistral AI | Microsoft |
| Noometry Index | 28.2 | 29.7 |
| Released | 2024-10-01 | 2024-04-23 |
| Weights | Open | Open |
| Context window | 262K | — |
| Max output | 262K | — |
| Input $ / M tokens | $0.15 | — |
| Output $ / M tokens | $0.15 | — |
| Results tracked | 17 | 13 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding phi-3-medium 14B leads
Ministral 8B: 35.0 (#230), phi-3-medium 14B: 36.8 (#201)
| Benchmark | Ministral 8B | phi-3-medium 14B |
|---|---|---|
| BigCodeBench Instruct | — | 37.6% |
| LMArena Coding | 1202 | — |
| BigCodeBench Complete | — | 48.7% |
Agentic & Tool Use Not comparable
Ministral 8B: 16.4 (#148), phi-3-medium 14B: —
| Benchmark | Ministral 8B | phi-3-medium 14B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 11.1% | — |
Reasoning Not comparable
Ministral 8B: 18.4 (#281), phi-3-medium 14B: —
| Benchmark | Ministral 8B | phi-3-medium 14B |
|---|---|---|
| LMArena Hard Prompts | 1191 | — |
| DTBench | 45.7% | — |
| Adversarial NLI | — | 55.8% |
| BIG-Bench Hard | — | 81.4% |
| Epoch Capabilities Index | — | 121.23 |
| HellaSwag | — | 82.4% |
| WinoGrande | — | 81.5% |
Math phi-3-medium 14B leads
Ministral 8B: 25.7 (#267), phi-3-medium 14B: 27.3 (#250)
| Benchmark | Ministral 8B | phi-3-medium 14B |
|---|---|---|
| MATH Level 5 | 14.9% | 17.6% |
| LMArena Math | 1188 | — |
Knowledge Ministral 8B leads
Ministral 8B: 12.6 (#297), phi-3-medium 14B: 9.1 (#306)
| Benchmark | Ministral 8B | phi-3-medium 14B |
|---|---|---|
| GPQA Diamond | 27.1% | 27.6% |
| Vectara Hallucination Rate | 7.4% | — |
| LMArena Expert | 1170 | — |
| ARC (AI2) Challenge | — | 91.6% |
| MMLU | — | 78% |
| OpenBookQA | — | 87.4% |
| TriviaQA | — | 73.9% |
Multilingual Not comparable
Ministral 8B: 35.1 (#247), phi-3-medium 14B: —
| Benchmark | Ministral 8B | phi-3-medium 14B |
|---|---|---|
| LMArena Non-English | 1165 | — |
| LMArena Chinese | 1193 | — |
| LMArena Russian | 1195 | — |
Instruction Following Not comparable
Ministral 8B: 60.5 (#250), phi-3-medium 14B: —
| Benchmark | Ministral 8B | phi-3-medium 14B |
|---|---|---|
| LMArena Instruction Following | 1161 | — |
Long Context Not comparable
Ministral 8B: 36.7 (#227), phi-3-medium 14B: —
| Benchmark | Ministral 8B | phi-3-medium 14B |
|---|---|---|
| LMArena Longer Query | 1212 | — |
Writing & Preference Not comparable
Ministral 8B: 39.6 (#246), phi-3-medium 14B: —
| Benchmark | Ministral 8B | phi-3-medium 14B |
|---|---|---|
| LMArena Text | 1191 | — |
| LMArena Creative Writing | 1175 | — |
| LMArena Multi-Turn | 1166 | — |
Frequently asked questions
Is Ministral 8B better than phi-3-medium 14B?
phi-3-medium 14B is the stronger model overall, scoring 29.7 to 28.2 on the Noometry Index.
Is Ministral 8B or phi-3-medium 14B better for coding?
phi-3-medium 14B scores higher on coding benchmarks: 36.8 versus 35.0 in the Noometry coding category.
How many benchmarks do Ministral 8B and phi-3-medium 14B share?
2 benchmarks have published results for both models. Ministral 8B has 17 scored results on Noometry and phi-3-medium 14B has 13.