North Small Translate vs Mistral Large 4
Side by side: price, limits, capabilities and real cost examples. Catalog data · checked Nov 16, 2026
Short answer
- Mistral Large 4 reads 32× more text per request (1.05M vs 32.8K tokens).
- Only Mistral Large 4 offers: tool calling, structured output, reasoning mode, image input, file input, prompt caching.
- A quality verdict needs a hands-on test. Neither model has completed our Hero suite yet, so this page compares facts, not output quality.
| North Small TranslateCohere | Mistral Large 4Mistral AI | |
|---|---|---|
| Input price / 1M tokens | $0 | $0.7 |
| Output price / 1M tokens | $0 | $2.1 |
| Context window | 32.8K | 1.05M |
| Max output per answer | 16K | 262.1K |
| Released | Sep 2026 | Oct 2026 |
| Tool calling | No | Yes |
| Structured output | No | Yes |
| Reasoning mode | No | Yes |
| Image input | No | Yes |
| File input | No | Yes |
| Prompt caching | No | Yes |
| What it costs in practice | ||
| 1,000 support replies1,000 chats with 800 input and 250 output tokens each | $0 | $1.09 |
| Summarize a 300-page reportabout 150,000 input tokens in one request, 2,000 tokens out | $0 | $0.109 |
| 10,000 product descriptions300 input and 200 output tokens per text | $0 | $6.3 |
| One day of agentic codingabout 3M input and 150K output tokens across many calls | $0 | $2.42 |
Green marks the better value in that row. Prices are first-party list prices in USD. Method.
