Model comparison
Llama 3.1-8B vs Mistral Medium
Mistral Medium is the stronger model overall, scoring 36.3 to 23.0 on the Noometry Index. Llama 3.1-8B costs 52× less per token, which makes it the better buy when Mistral Medium's lead doesn't matter for your workload.
Last verified . 26 shared benchmarks.
Meta
Mistral AISummary
- They share 26 benchmarks with published results for both. Llama 3.1-8B scores higher in 0 categories and Mistral Medium in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Mistral Medium leads 60.0 to 29.7.
- The biggest single-benchmark swing is MATH Level 5: 22.9% for Llama 3.1-8B and 81.6% for Mistral Medium.
- Llama 3.1-8B is cheaper at $0.05 / $0.08 per million input/output tokens, against $1.50 / $7.50 for Mistral Medium.
- Mistral Medium accepts more context: 262K tokens versus 128K.
Side by side
| Llama 3.1-8B | Mistral Medium | |
|---|---|---|
| Provider | Meta | Mistral AI |
| Noometry Index | 23.0 | 36.3 |
| Released | 2024-07-23 | 2023-12-11 |
| Weights | Open | Open |
| Context window | 128K | 262K |
| Max output | 4K | 262K |
| Input $ / M tokens | $0.05 | $1.50 |
| Output $ / M tokens | $0.08 | $7.50 |
| Results tracked | 43 | 36 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Mistral Medium leads
Llama 3.1-8B: 20.2 (#340), Mistral Medium: 34.2 (#243)
| Benchmark | Llama 3.1-8B | Mistral Medium |
|---|---|---|
| SciCode | 13.2% | 40.2% |
| WeirdML | 1.7% | 43.7% |
| LMArena Coding | 1195 | 1434 |
| FrontierCode | — | 8% |
| BigCodeBench Instruct | 32.8% | — |
| BigCodeBench Complete | 40.5% | — |
| ALE-Bench | — | 763.98 |
| HumanEval+ | 62.8% | — |
| MBPP+ | 55.6% | — |
Agentic & Tool Use Mistral Medium leads
Llama 3.1-8B: 22.5 (#131), Mistral Medium: 28.3 (#90)
| Benchmark | Llama 3.1-8B | Mistral Medium |
|---|---|---|
| Berkeley Function Calling Leaderboard | 25.8% | 37.7% |
| BALROG | 15.1% | — |
Reasoning Mistral Medium leads
Llama 3.1-8B: 14.9 (#321), Mistral Medium: 24.0 (#167)
| Benchmark | Llama 3.1-8B | Mistral Medium |
|---|---|---|
| CritPt | 0% | 0% |
| LMArena Hard Prompts | 1175 | 1426 |
| DTBench | 50.9% | 75.5% |
| LMCA | 5.4% | 26.1% |
| Kagi LLM Benchmark | — | 50% |
| Chess Puzzles | 0% | — |
| Surface Evolver Bench | — | 26.9% |
| Epoch Capabilities Index | 116.57 | — |
| PIQA | 81.2% | — |
Math Mistral Medium leads
Llama 3.1-8B: 10.2 (#317), Mistral Medium: 28.1 (#245)
| Benchmark | Llama 3.1-8B | Mistral Medium |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 1.7% | 32.2% |
| LMArena Math | 1179 | 1408 |
| MATH Level 5 | 22.9% | 81.6% |
| ProofBench | — | 9% |
| Omni-MATH | 13.7% | — |
| FrontierMath (Feb 2025 set) | — | 0.3% |
| GSM8K | 82.4% | — |
Knowledge Mistral Medium leads
Llama 3.1-8B: 8.0 (#307), Mistral Medium: 25.0 (#265)
| Benchmark | Llama 3.1-8B | Mistral Medium |
|---|---|---|
| GPQA Diamond | 27% | 59.5% |
| LMArena Expert | 1144 | 1408 |
| Humanity's Last Exam | — | 4.5% |
| MMLU-Pro | 40.6% | — |
| Vectara Hallucination Rate | — | 22.7% |
| GPQA (HELM) | 24.7% | — |
| BoolQ | 82.8% | — |
| MMLU | 56.1% | — |
Multimodal Not comparable
Llama 3.1-8B: —, Mistral Medium: 35.3 (#88)
| Benchmark | Llama 3.1-8B | Mistral Medium |
|---|---|---|
| LMArena Vision | — | 1172 |
Multilingual Mistral Medium leads
Llama 3.1-8B: 34.0 (#249), Mistral Medium: 52.1 (#91)
| Benchmark | Llama 3.1-8B | Mistral Medium |
|---|---|---|
| LMArena Non-English | 1148 | 1408 |
| LMArena Chinese | 1151 | 1447 |
| LMArena French | 1177 | 1459 |
| LMArena German | 1144 | 1432 |
| LMArena Japanese | 1061 | 1378 |
| LMArena Korean | 1053 | 1380 |
| LMArena Russian | 1158 | 1411 |
| LMArena Spanish | 1169 | 1433 |
Instruction Following Mistral Medium leads
Llama 3.1-8B: 58.9 (#258), Mistral Medium: 73.7 (#116)
| Benchmark | Llama 3.1-8B | Mistral Medium |
|---|---|---|
| LMArena Instruction Following | 1159 | 1398 |
| IFEval | 74.3% | — |
Long Context Mistral Medium leads
Llama 3.1-8B: 35.8 (#238), Mistral Medium: 42.9 (#114)
| Benchmark | Llama 3.1-8B | Mistral Medium |
|---|---|---|
| LMArena Longer Query | 1182 | 1406 |
Writing & Preference Mistral Medium leads
Llama 3.1-8B: 29.7 (#290), Mistral Medium: 60.0 (#103)
| Benchmark | Llama 3.1-8B | Mistral Medium |
|---|---|---|
| LMArena Text | 1187 | 1424 |
| LMArena Creative Writing | 1154 | 1391 |
| LMArena Multi-Turn | 1172 | 1418 |
| Short-Story Creative Writing | — | 77.3% |
| EQ-Bench Creative Writing | 713 | — |
| WildBench | 68.7% | — |
Frequently asked questions
Is Llama 3.1-8B better than Mistral Medium?
Mistral Medium is the stronger model overall, scoring 36.3 to 23.0 on the Noometry Index. Llama 3.1-8B costs 52× less per token, which makes it the better buy when Mistral Medium's lead doesn't matter for your workload.
Which is cheaper, Llama 3.1-8B or Mistral Medium?
Llama 3.1-8B is cheaper. It lists at $0.05 per million input tokens and $0.08 per million output tokens; Mistral Medium lists at $1.50 and $7.50.
Is Llama 3.1-8B or Mistral Medium better for coding?
Mistral Medium scores higher on coding benchmarks: 34.2 versus 20.2 in the Noometry coding category.
Which has the bigger context window?
Mistral Medium does, with 262K tokens against 128K.
How many benchmarks do Llama 3.1-8B and Mistral Medium share?
26 benchmarks have published results for both models. Llama 3.1-8B has 43 scored results on Noometry and Mistral Medium has 36.