High-speed compact model delivering low-latency inference at minimal compute cost Need a token count first? Use the Mistral token counter.
mistral-small-2603| Scenario | Input Tokens | Output Tokens | Uncached Cost | With Prompt Caching |
|---|---|---|---|---|
| Short Chat Query | 1,000 | 500 | $0.00045 | $0.00031 |
| Document Summarization | 10,000 | 2,000 | $0.00270 | $0.00135 |
| Codebase & Context Analysis | 100,000 | 20,000 | $0.0270 | $0.0135 |
| Batch Corpus Processing | 1,000,000 | 100,000 | $0.2100 | $0.0750 |
Tier 1 (free tier) reference limits — your account's actual quota may be higher.
Ideal for production workloads demanding budget capabilities, deep context depth (256k tokens), and reliability from Mistral AI. Excellent when predictable tokenomics and prompt caching support are paramount.
If your use-case requires sub-second streaming latency or ultra-high frequency classification at micro-cent pricing, consider lighter budget options such as Gemini Flash-Lite or Claude Haiku. For deep formal logic, consider dedicated reasoning models like o3.
For Mistral Small 4, 1 million input tokens costs $0.15, while 1 million output tokens costs $0.60. If using prompt caching, repetitive input prefixes are discounted to $0.015 per million.
Mistral Small 4 features a maximum context window of 262,144 tokens (~196,608 words), with a maximum output limit of 8,192 tokens per completion.
Mistral Small 4 utilizes the Mistral Tekken Tokenizer (131k vocabulary). Token counting on TokenMath runs client-side to ensure maximum privacy.
Yes. Mistral Small 4 supports prompt caching with a cached input rate of $0.015/1M (saving up to 90% on repeated input context).