Model comparison
Llama 3.1-8B vs Mistral Small
Mistral Small is the stronger model overall, scoring 33.4 to 23.0 on the Noometry Index. Llama 3.1-8B costs 4.6× less per token, which makes it the better buy when Mistral Small's lead doesn't matter for your workload.
Last verified . 28 shared benchmarks.
Meta
Mistral AISummary
- They share 28 benchmarks with published results for both. Llama 3.1-8B scores higher in 0 categories and Mistral Small in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Mistral Small leads 31.0 to 8.0.
- The biggest single-benchmark swing is MATH Level 5: 22.9% for Llama 3.1-8B and 46.8% for Mistral Small.
- Llama 3.1-8B is cheaper at $0.05 / $0.08 per million input/output tokens, against $0.15 / $0.60 for Mistral Small.
- Mistral Small accepts more context: 262K tokens versus 128K.
Side by side
| Llama 3.1-8B | Mistral Small | |
|---|---|---|
| Provider | Meta | Mistral AI |
| Noometry Index | 23.0 | 33.4 |
| Released | 2024-07-23 | 2024-02-26 |
| Weights | Open | Open |
| Context window | 128K | 262K |
| Max output | 4K | 256K |
| Input $ / M tokens | $0.05 | $0.15 |
| Output $ / M tokens | $0.08 | $0.60 |
| Results tracked | 43 | 39 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Mistral Small leads
Llama 3.1-8B: 20.2 (#340), Mistral Small: 34.0 (#247)
| Benchmark | Llama 3.1-8B | Mistral Small |
|---|---|---|
| SciCode | 13.2% | 26.5% |
| BigCodeBench Instruct | 32.8% | 36.1% |
| LMArena Coding | 1195 | 1362 |
| BigCodeBench Complete | 40.5% | 46.6% |
| WeirdML | 1.7% | — |
| LiveBench Coding | — | 36.2% |
| ALE-Bench | — | 497.62 |
| HumanEval+ | 62.8% | — |
| MBPP+ | 55.6% | — |
Agentic & Tool Use Mistral Small leads
Llama 3.1-8B: 22.5 (#131), Mistral Small: 28.1 (#93)
| Benchmark | Llama 3.1-8B | Mistral Small |
|---|---|---|
| Berkeley Function Calling Leaderboard | 25.8% | 37.1% |
| BALROG | 15.1% | — |
Reasoning Mistral Small leads
Llama 3.1-8B: 14.9 (#321), Mistral Small: 19.8 (#250)
| Benchmark | Llama 3.1-8B | Mistral Small |
|---|---|---|
| CritPt | 0% | 0% |
| LMArena Hard Prompts | 1175 | 1335 |
| DTBench | 50.9% | 70.9% |
| LMCA | 5.4% | 20.6% |
| Kagi LLM Benchmark | — | 37.8% |
| Chess Puzzles | 0% | — |
| LiveBench Reasoning | — | 44.8% |
| LiveBench Data Analysis | — | 53.7% |
| Epoch Capabilities Index | 116.57 | — |
| LiveBench | — | 44% |
| PIQA | 81.2% | — |
Math Mistral Small leads
Llama 3.1-8B: 10.2 (#317), Mistral Small: 16.4 (#293)
| Benchmark | Llama 3.1-8B | Mistral Small |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 1.7% | 5.8% |
| LMArena Math | 1179 | 1341 |
| MATH Level 5 | 22.9% | 46.8% |
| Omni-MATH | 13.7% | — |
| LiveBench Math | — | 39.9% |
| GSM8K | 82.4% | — |
Knowledge Mistral Small leads
Llama 3.1-8B: 8.0 (#307), Mistral Small: 31.0 (#222)
| Benchmark | Llama 3.1-8B | Mistral Small |
|---|---|---|
| GPQA Diamond | 27% | 47.5% |
| LMArena Expert | 1144 | 1291 |
| MMLU | 56.1% | 68.7% |
| MMLU-Pro | 40.6% | — |
| Vectara Hallucination Rate | — | 5.1% |
| GPQA (HELM) | 24.7% | — |
| BoolQ | 82.8% | — |
Multimodal Not comparable
Llama 3.1-8B: —, Mistral Small: 33.5 (#96)
| Benchmark | Llama 3.1-8B | Mistral Small |
|---|---|---|
| LMArena Vision | — | 1142 |
Multilingual Mistral Small leads
Llama 3.1-8B: 34.0 (#249), Mistral Small: 45.5 (#169)
| Benchmark | Llama 3.1-8B | Mistral Small |
|---|---|---|
| LMArena Non-English | 1148 | 1315 |
| LMArena Chinese | 1151 | 1340 |
| LMArena French | 1177 | 1337 |
| LMArena German | 1144 | 1340 |
| LMArena Japanese | 1061 | 1275 |
| LMArena Korean | 1053 | 1259 |
| LMArena Russian | 1158 | 1324 |
| LMArena Spanish | 1169 | 1346 |
Instruction Following Mistral Small leads
Llama 3.1-8B: 58.9 (#258), Mistral Small: 66.4 (#209)
| Benchmark | Llama 3.1-8B | Mistral Small |
|---|---|---|
| LMArena Instruction Following | 1159 | 1310 |
| LiveBench Instruction Following | — | 63.7% |
| IFEval | 74.3% | — |
Long Context Mistral Small leads
Llama 3.1-8B: 35.8 (#238), Mistral Small: 40.4 (#156)
| Benchmark | Llama 3.1-8B | Mistral Small |
|---|---|---|
| LMArena Longer Query | 1182 | 1327 |
Writing & Preference Mistral Small leads
Llama 3.1-8B: 29.7 (#290), Mistral Small: 52.5 (#171)
| Benchmark | Llama 3.1-8B | Mistral Small |
|---|---|---|
| LMArena Text | 1187 | 1338 |
| LMArena Creative Writing | 1154 | 1305 |
| LMArena Multi-Turn | 1172 | 1344 |
| EQ-Bench Creative Writing | 713 | — |
| WildBench | 68.7% | — |
| LiveBench Language | — | 30.5% |
Frequently asked questions
Is Llama 3.1-8B better than Mistral Small?
Mistral Small is the stronger model overall, scoring 33.4 to 23.0 on the Noometry Index. Llama 3.1-8B costs 4.6× less per token, which makes it the better buy when Mistral Small's lead doesn't matter for your workload.
Which is cheaper, Llama 3.1-8B or Mistral Small?
Llama 3.1-8B is cheaper. It lists at $0.05 per million input tokens and $0.08 per million output tokens; Mistral Small lists at $0.15 and $0.60.
Is Llama 3.1-8B or Mistral Small better for coding?
Mistral Small scores higher on coding benchmarks: 34.0 versus 20.2 in the Noometry coding category.
Which has the bigger context window?
Mistral Small does, with 262K tokens against 128K.
How many benchmarks do Llama 3.1-8B and Mistral Small share?
28 benchmarks have published results for both models. Llama 3.1-8B has 43 scored results on Noometry and Mistral Small has 39.