Model comparison
Llama 4 Maverick vs Mixtral 8x7B
Llama 4 Maverick is the stronger model overall, scoring 30.9 to 27.1 on the Noometry Index.
Last verified . 27 shared benchmarks.
Meta
Mistral AISummary
- They share 27 benchmarks with published results for both. Llama 4 Maverick scores higher in 5 categories and Mixtral 8x7B in 3 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Llama 4 Maverick leads 33.4 to 11.0.
- The biggest single-benchmark swing is MATH Level 5: 73% for Llama 4 Maverick and 10% for Mixtral 8x7B.
- Llama 4 Maverick is cheaper at $0.19 / $0.65 per million input/output tokens, against $0.70 / $0.70 for Mixtral 8x7B.
- Llama 4 Maverick accepts more context: 128K tokens versus 32K.
Side by side
| Llama 4 Maverick | Mixtral 8x7B | |
|---|---|---|
| Provider | Meta | Mistral AI |
| Noometry Index | 30.9 | 27.1 |
| Released | 2025-04-05 | 2023-12-11 |
| Weights | Open | Open |
| Context window | 128K | 32K |
| Max output | 4K | 32K |
| Input $ / M tokens | $0.19 | $0.70 |
| Output $ / M tokens | $0.65 | $0.70 |
| Results tracked | 54 | 38 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Mixtral 8x7B leads
Llama 4 Maverick: 26.6 (#324), Mixtral 8x7B: 32.8 (#269)
| Benchmark | Llama 4 Maverick | Mixtral 8x7B |
|---|---|---|
| LMArena Coding | 1302 | 1126 |
| 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 | — |
| HumanEval+ | — | 39.6% |
| MBPP+ | — | 49.7% |
Agentic & Tool Use Not comparable
Llama 4 Maverick: 28.2 (#91), Mixtral 8x7B: —
| Benchmark | Llama 4 Maverick | Mixtral 8x7B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 37.3% | — |
Reasoning Mixtral 8x7B leads
Llama 4 Maverick: 10.1 (#342), Mixtral 8x7B: 18.2 (#285)
| Benchmark | Llama 4 Maverick | Mixtral 8x7B |
|---|---|---|
| LMArena Hard Prompts | 1281 | 1115 |
| DTBench | 61.9% | 49.6% |
| Epoch Capabilities Index | 132.2 | 118.47 |
| ForecastBench | 57.5 | 56.3 |
| 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% | — |
| Adversarial NLI | — | 55.2% |
| HellaSwag | — | 86.7% |
| PIQA | — | 83.6% |
| WinoGrande | — | 77.2% |
Math Llama 4 Maverick leads
Llama 4 Maverick: 26.0 (#262), Mixtral 8x7B: 18.8 (#289)
| Benchmark | Llama 4 Maverick | Mixtral 8x7B |
|---|---|---|
| Omni-MATH | 42.2% | 10.5% |
| LMArena Math | 1299 | 1147 |
| MATH Level 5 | 73% | 10% |
| OTIS Mock AIME 2024-2025 | 20.6% | — |
| FrontierMath (Feb 2025 set) | 0.7% | — |
| GSM8K | — | 74.4% |
Knowledge Llama 4 Maverick leads
Llama 4 Maverick: 33.4 (#204), Mixtral 8x7B: 11.0 (#301)
| Benchmark | Llama 4 Maverick | Mixtral 8x7B |
|---|---|---|
| GPQA Diamond | 67% | 30.6% |
| MMLU-Pro | 81% | 33.5% |
| GPQA (HELM) | 65% | 29.6% |
| LMArena Expert | 1259 | 1088 |
| Humanity's Last Exam | 5.7% | — |
| Confabulations | 22.6% | — |
| Vectara Hallucination Rate | 8.2% | — |
| ARC (AI2) Challenge | — | 87.3% |
| MMLU | — | 70.6% |
| OpenBookQA | — | 85.8% |
| TriviaQA | — | 82.2% |
Multimodal Not comparable
Llama 4 Maverick: 31.6 (#105), Mixtral 8x7B: —
| Benchmark | Llama 4 Maverick | Mixtral 8x7B |
|---|---|---|
| LMArena Vision | 1142 | — |
| GeoBench | 52% | — |
| SpatialViz-Bench | 31.8% | — |
Multilingual Llama 4 Maverick leads
Llama 4 Maverick: 42.2 (#195), Mixtral 8x7B: 29.6 (#266)
| Benchmark | Llama 4 Maverick | Mixtral 8x7B |
|---|---|---|
| LMArena Non-English | 1269 | 1077 |
| LMArena Chinese | 1277 | 1055 |
| LMArena French | 1259 | 1166 |
| LMArena German | 1291 | 1114 |
| LMArena Japanese | 1207 | 931 |
| LMArena Korean | 1203 | 968 |
| LMArena Russian | 1286 | 1090 |
| LMArena Spanish | 1293 | 1111 |
Instruction Following Llama 4 Maverick leads
Llama 4 Maverick: 71.7 (#146), Mixtral 8x7B: 51.0 (#297)
| Benchmark | Llama 4 Maverick | Mixtral 8x7B |
|---|---|---|
| IFEval | 90.8% | 57.5% |
| LMArena Instruction Following | 1267 | 1109 |
Long Context Mixtral 8x7B leads
Llama 4 Maverick: 31.4 (#279), Mixtral 8x7B: 33.4 (#260)
| Benchmark | Llama 4 Maverick | Mixtral 8x7B |
|---|---|---|
| LMArena Longer Query | 1280 | 1103 |
| Fiction.LiveBench | 46.2% | — |
Writing & Preference Llama 4 Maverick leads
Llama 4 Maverick: 38.8 (#252), Mixtral 8x7B: 34.2 (#270)
| Benchmark | Llama 4 Maverick | Mixtral 8x7B |
|---|---|---|
| LMArena Text | 1287 | 1132 |
| LMArena Creative Writing | 1267 | 1109 |
| WildBench | 80% | 67.3% |
| LMArena Multi-Turn | 1289 | 1115 |
| Short-Story Creative Writing | 62% | — |
| EQ-Bench Creative Writing | 860 | — |
Frequently asked questions
Is Llama 4 Maverick better than Mixtral 8x7B?
Llama 4 Maverick is the stronger model overall, scoring 30.9 to 27.1 on the Noometry Index.
Which is cheaper, Llama 4 Maverick or Mixtral 8x7B?
Llama 4 Maverick is cheaper. It lists at $0.19 per million input tokens and $0.65 per million output tokens; Mixtral 8x7B lists at $0.70 and $0.70.
Is Llama 4 Maverick or Mixtral 8x7B better for coding?
Mixtral 8x7B scores higher on coding benchmarks: 32.8 versus 26.6 in the Noometry coding category.
Which has the bigger context window?
Llama 4 Maverick does, with 128K tokens against 32K.
How many benchmarks do Llama 4 Maverick and Mixtral 8x7B share?
27 benchmarks have published results for both models. Llama 4 Maverick has 54 scored results on Noometry and Mixtral 8x7B has 38.