# Llama 3-8B vs Magistral Small

> Magistral Small is the stronger model overall, scoring 29.9 to 25.3 on the Noometry Index. Llama 3-8B costs 5.4× less per token, which makes it the better buy when Magistral Small's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/llama-3-8b-vs-magistral-small
- Last updated: 2026-10-10
- Shared benchmarks: 5

## Summary

- They share 5 benchmarks with published results for both. Llama 3-8B scores higher in 1 category and Magistral Small in 3 categories; 4 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Magistral Small leads 30.9 to 7.8.
- The biggest single-benchmark swing is GPQA Diamond: 26.1% for Llama 3-8B and 56.1% for Magistral Small.
- Llama 3-8B is cheaper at $0.14 / $0.14 per million input/output tokens, against $0.50 / $1.50 for Magistral Small.
- Magistral Small accepts more context: 128K tokens versus 8K.

## Snapshot

| | Llama 3-8B | Magistral Small |
|---|---|---|
| Provider | Meta | Mistral AI |
| Noometry Index | 25.3 | 29.9 |
| Rank | 338 | 289 |
| Context | 8K | 128K |
| Input $/M | $0.14 | $0.50 |
| Output $/M | $0.14 | $1.50 |
| Weights | Open | Open |

## Coding

- Llama 3-8B: 30.5 (#285)
- Magistral Small: 37.8 (#172)

| Benchmark | Llama 3-8B | Magistral Small |
|---|---|---|
| SciCode | — | 35.2% |
| BigCodeBench Instruct | 31.9% | — |
| LMArena Coding | 1152 | — |
| BigCodeBench Complete | 36.9% | — |
| HumanEval+ | 56.7% | — |
| MBPP+ | 54.8% | — |

## Reasoning

- Llama 3-8B: 13.8 (#321)
- Magistral Small: 6.9 (#345)

| Benchmark | Llama 3-8B | Magistral Small |
|---|---|---|
| Chess Puzzles | 0% | 3% |
| DTBench | 43.9% | 61.3% |
| Epoch Capabilities Index | 116.33 | 133.19 |
| ARC-AGI-2 | — | 0% |
| Kagi LLM Benchmark | — | 6.3% |
| ARC-AGI-1 | — | 5% |
| CritPt | — | 0.3% |
| LMArena Hard Prompts | 1133 | — |
| Adversarial NLI | 57.3% | — |
| ForecastBench | 58.6 | — |
| WinoGrande | 75.7% | — |

## Math

- Llama 3-8B: 8.3 (#317)
- Magistral Small: 24.9 (#253)

| Benchmark | Llama 3-8B | Magistral Small |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 1.9% | 30% |
| LMArena Math | 1151 | — |
| MATH Level 5 | 6.1% | — |

## Knowledge

- Llama 3-8B: 7.8 (#303)
- Magistral Small: 30.9 (#215)

| Benchmark | Llama 3-8B | Magistral Small |
|---|---|---|
| GPQA Diamond | 26.1% | 56.1% |
| LMArena Expert | 1113 | — |
| ARC (AI2) Challenge | 82.8% | — |
| MMLU | 68.8% | — |
| OpenBookQA | 82.6% | — |
| TriviaQA | 67.7% | — |

## Multilingual

- Llama 3-8B: 31.0 (#256)
- Magistral Small: —

| Benchmark | Llama 3-8B | Magistral Small |
|---|---|---|
| LMArena Non-English | 1098 | — |
| LMArena Chinese | 1076 | — |
| LMArena French | 1158 | — |
| LMArena German | 1104 | — |
| LMArena Japanese | 967 | — |
| LMArena Korean | 1003 | — |
| LMArena Russian | 1109 | — |
| LMArena Spanish | 1172 | — |

## Instruction Following

- Llama 3-8B: 58.7 (#255)
- Magistral Small: —

| Benchmark | Llama 3-8B | Magistral Small |
|---|---|---|
| LMArena Instruction Following | 1127 | — |

## Long Context

- Llama 3-8B: 34.2 (#246)
- Magistral Small: —

| Benchmark | Llama 3-8B | Magistral Small |
|---|---|---|
| LMArena Longer Query | 1128 | — |

## Writing & Preference

- Llama 3-8B: 37.7 (#251)
- Magistral Small: —

| Benchmark | Llama 3-8B | Magistral Small |
|---|---|---|
| LMArena Text | 1166 | — |
| LMArena Creative Writing | 1150 | — |
| LMArena Multi-Turn | 1152 | — |

## FAQ

### Is Llama 3-8B better than Magistral Small?

Magistral Small is the stronger model overall, scoring 29.9 to 25.3 on the Noometry Index. Llama 3-8B costs 5.4× less per token, which makes it the better buy when Magistral Small's lead doesn't matter for your workload.

### Which is cheaper, Llama 3-8B or Magistral Small?

Llama 3-8B is cheaper. It lists at $0.14 per million input tokens and $0.14 per million output tokens; Magistral Small lists at $0.50 and $1.50.

### Is Llama 3-8B or Magistral Small better for coding?

Magistral Small scores higher on coding benchmarks: 37.8 versus 30.5 in the Noometry coding category.

### Which has the bigger context window?

Magistral Small does, with 128K tokens against 8K.

### How many benchmarks do Llama 3-8B and Magistral Small share?

5 benchmarks have published results for both models. Llama 3-8B has 34 scored results on Noometry and Magistral Small has 10.