Model comparison
Codestral vs Phi-4
Codestral and Phi-4 score almost the same on the Noometry Index (30.6 vs 31.2), so choose on price, context window or the category you care about most.
Last verified . 2 shared benchmarks.
Mistral AI
MicrosoftSummary
- They share 2 benchmarks with published results for both. Codestral scores higher in 1 category and Phi-4 in 1 category; 2 gaps are clear of the uncertainty.
- The widest gap is in coding, where Phi-4 leads 34.4 to 27.3.
- Phi-4 is cheaper at $0.07 / $0.14 per million input/output tokens, against $0.30 / $0.90 for Codestral.
- Codestral accepts more context: 256K tokens versus 128K.
- Phi-4 has downloadable open weights; the other is API-only.
Side by side
| Codestral | Phi-4 | |
|---|---|---|
| Provider | Mistral AI | Microsoft |
| Noometry Index | 30.6 | 31.2 |
| Released | 2024-05-29 | 2024-12-11 |
| Weights | Proprietary | Open |
| Context window | 256K | 128K |
| Max output | 8K | 4K |
| Input $ / M tokens | $0.30 | $0.07 |
| Output $ / M tokens | $0.90 | $0.14 |
| Results tracked | 7 | 37 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Phi-4 leads
Codestral: 27.3 (#321), Phi-4: 34.4 (#239)
| Benchmark | Codestral | Phi-4 |
|---|---|---|
| BigCodeBench Instruct | 41.8% | 45.5% |
| BigCodeBench Complete | 52.5% | 55.4% |
| Aider Polyglot | 11.1% | — |
| LiveBench Coding | — | 30.7% |
| LMArena Coding | — | 1231 |
| ALE-Bench | 137.78 | — |
| HumanEval+ | 73.8% | — |
| MBPP+ | 61.9% | — |
Agentic & Tool Use Not comparable
Codestral: —, Phi-4: 22.8 (#128)
| Benchmark | Codestral | Phi-4 |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 28.8% |
| BALROG | — | 11.6% |
Reasoning Codestral leads
Codestral: 19.8 (#251), Phi-4: 17.7 (#291)
| Benchmark | Codestral | Phi-4 |
|---|---|---|
| Kagi LLM Benchmark | 32.5% | — |
| Chess Puzzles | — | 1% |
| LiveBench Reasoning | — | 47.8% |
| LMArena Hard Prompts | — | 1220 |
| LiveBench Data Analysis | — | 45.2% |
| Epoch Capabilities Index | — | 130.42 |
| LiveBench | — | 41.6% |
Math Not comparable
Codestral: —, Phi-4: 20.8 (#285)
| Benchmark | Codestral | Phi-4 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | — | 13.8% |
| LiveBench Math | — | 42% |
| LMArena Math | — | 1246 |
| MATH Level 5 | — | 64.9% |
Knowledge Not comparable
Codestral: —, Phi-4: 32.6 (#209)
| Benchmark | Codestral | Phi-4 |
|---|---|---|
| GPQA Diamond | — | 56.1% |
| Confabulations | — | 29.4% |
| Vectara Hallucination Rate | — | 3.7% |
| LMArena Expert | — | 1203 |
| MMLU | — | 84.8% |
Multilingual Not comparable
Codestral: —, Phi-4: 37.2 (#237)
| Benchmark | Codestral | Phi-4 |
|---|---|---|
| LMArena Non-English | — | 1197 |
| LMArena Chinese | — | 1212 |
| LMArena French | — | 1224 |
| LMArena German | — | 1222 |
| LMArena Japanese | — | 1158 |
| LMArena Korean | — | 1151 |
| LMArena Russian | — | 1209 |
| LMArena Spanish | — | 1234 |
Instruction Following Not comparable
Codestral: —, Phi-4: 60.4 (#251)
| Benchmark | Codestral | Phi-4 |
|---|---|---|
| LiveBench Instruction Following | — | 58.4% |
| LMArena Instruction Following | — | 1201 |
Long Context Not comparable
Codestral: —, Phi-4: 36.9 (#226)
| Benchmark | Codestral | Phi-4 |
|---|---|---|
| LMArena Longer Query | — | 1217 |
Writing & Preference Not comparable
Codestral: —, Phi-4: 40.5 (#244)
| Benchmark | Codestral | Phi-4 |
|---|---|---|
| LMArena Text | — | 1217 |
| LMArena Creative Writing | — | 1182 |
| Short-Story Creative Writing | — | 62.6% |
| LMArena Multi-Turn | — | 1206 |
| LiveBench Language | — | 25.6% |
Frequently asked questions
Is Codestral better than Phi-4?
Codestral and Phi-4 score almost the same on the Noometry Index (30.6 vs 31.2), so choose on price, context window or the category you care about most.
Which is cheaper, Codestral or Phi-4?
Phi-4 is cheaper. It lists at $0.07 per million input tokens and $0.14 per million output tokens; Codestral lists at $0.30 and $0.90.
Is Codestral or Phi-4 better for coding?
Phi-4 scores higher on coding benchmarks: 34.4 versus 27.3 in the Noometry coding category.
Which has the bigger context window?
Codestral does, with 256K tokens against 128K.
How many benchmarks do Codestral and Phi-4 share?
2 benchmarks have published results for both models. Codestral has 7 scored results on Noometry and Phi-4 has 37.