Meta-Llama-3-8B-Instruct vs Mistral Small (latest)
Mistral Small (latest) is cheaper on output tokens, while Mistral Small (latest) offers a larger context window. Choose Meta-Llama-3-8B-Instruct or Mistral Small (latest) based on the trade-off between cost, context, and the benchmarks that matter for your use case.
| Spec | Meta-Llama-3-8B-Instruct | Mistral Small (latest) |
|---|---|---|
| Provider | Azure | Mistral |
| Input / 1M tokens | $0.30 | $0.15 |
| Output / 1M tokens | $0.61 | $0.60 |
| Context window | 8K | 262K |
| Parameters | — | — |
| Open weights | Yes | Yes |
| Released | Apr 2024 | Mar 2026 |
| GPQA Diamond score | 26.1 | 47.5 |
| LMArena (Chatbot Arena) Elo score | — | 1338 |
| LMArena Vision Elo score | — | 1142 |
FAQ
Is Meta-Llama-3-8B-Instruct or Mistral Small (latest) cheaper?
Mistral Small (latest) is cheaper on output tokens ($0.60 vs $0.61 per 1M).
Which has the larger context window, Meta-Llama-3-8B-Instruct or Mistral Small (latest)?
Mistral Small (latest) has the larger context window (262K tokens).
Pricing is indicative — confirm with the provider before production use. Highlighted values indicate the better figure for that row.