Model comparison
Llama 4 Maverick vs Magistral Medium
Magistral Medium is the stronger model overall, scoring 35.2 to 30.9 on the Noometry Index. Llama 4 Maverick costs 9.1× less per token, which makes it the better buy when Magistral Medium's lead doesn't matter for your workload.
Last verified . 22 shared benchmarks.
Meta
Mistral AISummary
- They share 22 benchmarks with published results for both. Llama 4 Maverick scores higher in 3 categories and Magistral Medium in 5 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in coding, where Magistral Medium leads 39.1 to 26.6.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 55.9% for Llama 4 Maverick and 16.2% for Magistral Medium.
- Llama 4 Maverick is cheaper at $0.19 / $0.65 per million input/output tokens, against $2 / $5 for Magistral Medium.
- Magistral Medium accepts more context: 262K tokens versus 128K.
Side by side
| Llama 4 Maverick | Magistral Medium | |
|---|---|---|
| Provider | Meta | Mistral AI |
| Noometry Index | 30.9 | 35.2 |
| Released | 2025-04-05 | 2025-03-17 |
| Weights | Open | Open |
| Context window | 128K | 262K |
| Max output | 4K | 16K |
| Input $ / M tokens | $0.19 | $2 |
| Output $ / M tokens | $0.65 | $5 |
| Results tracked | 54 | 22 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Magistral Medium leads
Llama 4 Maverick: 26.6 (#324), Magistral Medium: 39.1 (#161)
| Benchmark | Llama 4 Maverick | Magistral Medium |
|---|---|---|
| SciCode | 33.1% | 39.2% |
| LMArena Coding | 1302 | 1319 |
| SWE-bench Verified (bash only) | 21% | — |
| Aider Polyglot | 15.6% | — |
| WeirdML | 24.5% | — |
| BigCodeBench Instruct | 49.7% | — |
| BigCodeBench Complete | 61.4% | — |
| ALE-Bench | 172.97 | — |
Agentic & Tool Use Not comparable
Llama 4 Maverick: 28.2 (#91), Magistral Medium: —
| Benchmark | Llama 4 Maverick | Magistral Medium |
|---|---|---|
| Berkeley Function Calling Leaderboard | 37.3% | — |
Reasoning Llama 4 Maverick leads
Llama 4 Maverick: 10.1 (#342), Magistral Medium: 8.6 (#348)
| Benchmark | Llama 4 Maverick | Magistral Medium |
|---|---|---|
| ARC-AGI-2 | 0% | 0% |
| Kagi LLM Benchmark | 55.9% | 16.2% |
| ARC-AGI-1 | 4.4% | 6.1% |
| CritPt | 0% | 0.3% |
| LMArena Hard Prompts | 1281 | 1267 |
| SimpleBench | 27.7% | — |
| NYT Connections (extended) | 8% | — |
| EnigmaEval | 0.6% | — |
| DTBench | 61.9% | — |
| LMCA | 15.9% | — |
| Epoch Capabilities Index | 132.2 | — |
| ForecastBench | 57.5 | — |
Math Magistral Medium leads
Llama 4 Maverick: 26.0 (#262), Magistral Medium: 35.1 (#189)
| Benchmark | Llama 4 Maverick | Magistral Medium |
|---|---|---|
| LMArena Math | 1299 | 1250 |
| OTIS Mock AIME 2024-2025 | 20.6% | — |
| Omni-MATH | 42.2% | — |
| MATH Level 5 | 73% | — |
| FrontierMath (Feb 2025 set) | 0.7% | — |
Knowledge Too close to call
Llama 4 Maverick: 33.4 (#204), Magistral Medium: 33.5 (#202)
| Benchmark | Llama 4 Maverick | Magistral Medium |
|---|---|---|
| LMArena Expert | 1259 | 1223 |
| GPQA Diamond | 67% | — |
| Humanity's Last Exam | 5.7% | — |
| MMLU-Pro | 81% | — |
| Confabulations | 22.6% | — |
| Vectara Hallucination Rate | 8.2% | — |
| GPQA (HELM) | 65% | — |
Multimodal Not comparable
Llama 4 Maverick: 31.6 (#105), Magistral Medium: —
| Benchmark | Llama 4 Maverick | Magistral Medium |
|---|---|---|
| LMArena Vision | 1142 | — |
| GeoBench | 52% | — |
| SpatialViz-Bench | 31.8% | — |
Multilingual Llama 4 Maverick leads
Llama 4 Maverick: 42.2 (#195), Magistral Medium: 39.6 (#224)
| Benchmark | Llama 4 Maverick | Magistral Medium |
|---|---|---|
| LMArena Non-English | 1269 | 1232 |
| LMArena Chinese | 1277 | 1227 |
| LMArena French | 1259 | 1267 |
| LMArena German | 1291 | 1248 |
| LMArena Japanese | 1207 | 1175 |
| LMArena Korean | 1203 | 1125 |
| LMArena Russian | 1286 | 1224 |
| LMArena Spanish | 1293 | 1271 |
Instruction Following Llama 4 Maverick leads
Llama 4 Maverick: 71.7 (#146), Magistral Medium: 66.0 (#211)
| Benchmark | Llama 4 Maverick | Magistral Medium |
|---|---|---|
| LMArena Instruction Following | 1267 | 1254 |
| IFEval | 90.8% | — |
Long Context Magistral Medium leads
Llama 4 Maverick: 31.4 (#279), Magistral Medium: 39.3 (#183)
| Benchmark | Llama 4 Maverick | Magistral Medium |
|---|---|---|
| LMArena Longer Query | 1280 | 1295 |
| Fiction.LiveBench | 46.2% | — |
Writing & Preference Magistral Medium leads
Llama 4 Maverick: 38.8 (#252), Magistral Medium: 46.3 (#219)
| Benchmark | Llama 4 Maverick | Magistral Medium |
|---|---|---|
| LMArena Text | 1287 | 1255 |
| LMArena Creative Writing | 1267 | 1245 |
| LMArena Multi-Turn | 1289 | 1275 |
| Short-Story Creative Writing | 62% | — |
| EQ-Bench Creative Writing | 860 | — |
| WildBench | 80% | — |
Frequently asked questions
Is Llama 4 Maverick better than Magistral Medium?
Magistral Medium is the stronger model overall, scoring 35.2 to 30.9 on the Noometry Index. Llama 4 Maverick costs 9.1× less per token, which makes it the better buy when Magistral Medium's lead doesn't matter for your workload.
Which is cheaper, Llama 4 Maverick or Magistral Medium?
Llama 4 Maverick is cheaper. It lists at $0.19 per million input tokens and $0.65 per million output tokens; Magistral Medium lists at $2 and $5.
Is Llama 4 Maverick or Magistral Medium better for coding?
Magistral Medium scores higher on coding benchmarks: 39.1 versus 26.6 in the Noometry coding category.
Which has the bigger context window?
Magistral Medium does, with 262K tokens against 128K.
How many benchmarks do Llama 4 Maverick and Magistral Medium share?
22 benchmarks have published results for both models. Llama 4 Maverick has 54 scored results on Noometry and Magistral Medium has 22.