Mistral Small 4
Mistral Small 4 unifies Instruct, Magistral, and Devstral capabilities into a single 119B MoE architecture activating just 6.5B parameters. It offers configurable reasoning effort and native multimodality.
Leaderboards
Average Score combining domain-specific Autobench scores; Higher is better
Performance vs. Industry Average
Intelligence
Mistral Small 4 is of lower intelligence compared to average (2.9), with an intelligence score of 2.7.
Price
Mistral Small 4 is cheaper compared to average ($0.75 per 1M Tokens) with a price of $0.05 per 1M Tokens.
Latency
Mistral Small 4 has a lower average latency compared to average (44.25s), with an average latency of 10.56s.
P99 Latency
Mistral Small 4 has a lower P99 latency compared to average (126.46s), taking 41.25s to receive the first token at P99 (TTFT).
Context Window
Mistral Small 4 has a smaller context window than average (406k tokens), with a context window of 262k tokens.