Model comparison
Llama 4 Maverick vs Ministral 8B
Llama 4 Maverick is the stronger model overall, scoring 30.9 to 28.2 on the Noometry Index. Ministral 8B costs 2.0× less per token, which makes it the better buy when Llama 4 Maverick's lead doesn't matter for your workload.
Last verified . 17 shared benchmarks.
Meta
Mistral AISummary
- They share 17 benchmarks with published results for both. Llama 4 Maverick scores higher in 5 categories and Ministral 8B in 4 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Llama 4 Maverick leads 33.4 to 12.6.
- The biggest single-benchmark swing is MATH Level 5: 73% for Llama 4 Maverick and 14.9% for Ministral 8B.
- Ministral 8B is cheaper at $0.15 / $0.15 per million input/output tokens, against $0.19 / $0.65 for Llama 4 Maverick.
- Ministral 8B accepts more context: 262K tokens versus 128K.
Side by side
| Llama 4 Maverick | Ministral 8B | |
|---|---|---|
| Provider | Meta | Mistral AI |
| Noometry Index | 30.9 | 28.2 |
| Released | 2025-04-05 | 2024-10-01 |
| Weights | Open | Open |
| Context window | 128K | 262K |
| Max output | 4K | 262K |
| Input $ / M tokens | $0.19 | $0.15 |
| Output $ / M tokens | $0.65 | $0.15 |
| Results tracked | 54 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Ministral 8B leads
Llama 4 Maverick: 26.6 (#324), Ministral 8B: 35.0 (#230)
| Benchmark | Llama 4 Maverick | Ministral 8B |
|---|---|---|
| LMArena Coding | 1302 | 1202 |
| SWE-bench Verified (bash only) | 21% | — |
| Aider Polyglot | 15.6% | — |
| SciCode | 33.1% | — |
| WeirdML | 24.5% | — |
| BigCodeBench Instruct | 49.7% | — |
| BigCodeBench Complete | 61.4% | — |
| ALE-Bench | 172.97 | — |
Agentic & Tool Use Llama 4 Maverick leads
Llama 4 Maverick: 28.2 (#91), Ministral 8B: 16.4 (#148)
| Benchmark | Llama 4 Maverick | Ministral 8B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 37.3% | 11.1% |
Reasoning Ministral 8B leads
Llama 4 Maverick: 10.1 (#342), Ministral 8B: 18.4 (#281)
| Benchmark | Llama 4 Maverick | Ministral 8B |
|---|---|---|
| LMArena Hard Prompts | 1281 | 1191 |
| DTBench | 61.9% | 45.7% |
| ARC-AGI-2 | 0% | — |
| SimpleBench | 27.7% | — |
| Kagi LLM Benchmark | 55.9% | — |
| NYT Connections (extended) | 8% | — |
| ARC-AGI-1 | 4.4% | — |
| CritPt | 0% | — |
| EnigmaEval | 0.6% | — |
| LMCA | 15.9% | — |
| Epoch Capabilities Index | 132.2 | — |
| ForecastBench | 57.5 | — |
Math Too close to call
Llama 4 Maverick: 26.0 (#262), Ministral 8B: 25.7 (#267)
| Benchmark | Llama 4 Maverick | Ministral 8B |
|---|---|---|
| LMArena Math | 1299 | 1188 |
| MATH Level 5 | 73% | 14.9% |
| OTIS Mock AIME 2024-2025 | 20.6% | — |
| Omni-MATH | 42.2% | — |
| FrontierMath (Feb 2025 set) | 0.7% | — |
Knowledge Llama 4 Maverick leads
Llama 4 Maverick: 33.4 (#204), Ministral 8B: 12.6 (#297)
| Benchmark | Llama 4 Maverick | Ministral 8B |
|---|---|---|
| GPQA Diamond | 67% | 27.1% |
| Vectara Hallucination Rate | 8.2% | 7.4% |
| LMArena Expert | 1259 | 1170 |
| Humanity's Last Exam | 5.7% | — |
| MMLU-Pro | 81% | — |
| Confabulations | 22.6% | — |
| GPQA (HELM) | 65% | — |
Multimodal Not comparable
Llama 4 Maverick: 31.6 (#105), Ministral 8B: —
| Benchmark | Llama 4 Maverick | Ministral 8B |
|---|---|---|
| LMArena Vision | 1142 | — |
| GeoBench | 52% | — |
| SpatialViz-Bench | 31.8% | — |
Multilingual Llama 4 Maverick leads
Llama 4 Maverick: 42.2 (#195), Ministral 8B: 35.1 (#247)
| Benchmark | Llama 4 Maverick | Ministral 8B |
|---|---|---|
| LMArena Non-English | 1269 | 1165 |
| LMArena Chinese | 1277 | 1193 |
| LMArena Russian | 1286 | 1195 |
| LMArena French | 1259 | — |
| LMArena German | 1291 | — |
| LMArena Japanese | 1207 | — |
| LMArena Korean | 1203 | — |
| LMArena Spanish | 1293 | — |
Instruction Following Llama 4 Maverick leads
Llama 4 Maverick: 71.7 (#146), Ministral 8B: 60.5 (#250)
| Benchmark | Llama 4 Maverick | Ministral 8B |
|---|---|---|
| LMArena Instruction Following | 1267 | 1161 |
| IFEval | 90.8% | — |
Long Context Ministral 8B leads
Llama 4 Maverick: 31.4 (#279), Ministral 8B: 36.7 (#227)
| Benchmark | Llama 4 Maverick | Ministral 8B |
|---|---|---|
| LMArena Longer Query | 1280 | 1212 |
| Fiction.LiveBench | 46.2% | — |
Writing & Preference Too close to call
Llama 4 Maverick: 38.8 (#252), Ministral 8B: 39.6 (#246)
| Benchmark | Llama 4 Maverick | Ministral 8B |
|---|---|---|
| LMArena Text | 1287 | 1191 |
| LMArena Creative Writing | 1267 | 1175 |
| LMArena Multi-Turn | 1289 | 1166 |
| Short-Story Creative Writing | 62% | — |
| EQ-Bench Creative Writing | 860 | — |
| WildBench | 80% | — |
Frequently asked questions
Is Llama 4 Maverick better than Ministral 8B?
Llama 4 Maverick is the stronger model overall, scoring 30.9 to 28.2 on the Noometry Index. Ministral 8B costs 2.0× less per token, which makes it the better buy when Llama 4 Maverick's lead doesn't matter for your workload.
Which is cheaper, Llama 4 Maverick or Ministral 8B?
Ministral 8B is cheaper. It lists at $0.15 per million input tokens and $0.15 per million output tokens; Llama 4 Maverick lists at $0.19 and $0.65.
Is Llama 4 Maverick or Ministral 8B better for coding?
Ministral 8B scores higher on coding benchmarks: 35.0 versus 26.6 in the Noometry coding category.
Which has the bigger context window?
Ministral 8B does, with 262K tokens against 128K.
How many benchmarks do Llama 4 Maverick and Ministral 8B share?
17 benchmarks have published results for both models. Llama 4 Maverick has 54 scored results on Noometry and Ministral 8B has 17.