Live · serverless · 3 regions
Devstral Small 2
by Mistral AI
A 24B model tuned for software engineering agents that open files, run commands and fix failing tests.
- code
- FP8
Serverless priceUSD
- Input
- $0.09per 1M tokens
- Output
- $0.30per 1M tokens
Billed per token, no minimums.
Reference
Specifications
- Model ID
- Endpoint
- Publisher
- Tasks
- Served context
- Quantization
- Decode speed
- p50 first token
- Regions
- Dedicated
Last 24 hours
Latency and speed
Median of each 15-minute step on the shared pool, measured at the API edge. Time to first token includes queueing; decode speed is for one request.
Time to first token · ms
120 msmedian
p95 148 ms
Decode speed · tok/s
160 tok/smedian
p5 143 tok/s
Show the data
| Step | Time to first token | Decode speed |
|---|---|---|
| 15m ago | 117 ms | 165 tok/s |
| 30m ago | 124 ms | 158 tok/s |
| 45m ago | 126 ms | 159 tok/s |
| 1h ago | 128 ms | 153 tok/s |
| 1h 15m ago | 121 ms | 160 tok/s |
| 1h 30m ago | 184 ms | 120 tok/s |
| 1h 45m ago | 126 ms | 159 tok/s |
| 2h ago | 122 ms | 163 tok/s |
| 2h 15m ago | 126 ms | 153 tok/s |
| 2h 30m ago | 124 ms | 156 tok/s |
| 2h 45m ago | 119 ms | 164 tok/s |
| 3h ago | 124 ms | 153 tok/s |
| 3h 15m ago | 130 ms | 153 tok/s |
| 3h 30m ago | 122 ms | 156 tok/s |
| 3h 45m ago | 121 ms | 158 tok/s |
| 4h ago | 126 ms | 154 tok/s |
| 4h 15m ago | 118 ms | 158 tok/s |
| 4h 30m ago | 123 ms | 159 tok/s |
| 4h 45m ago | 122 ms | 155 tok/s |
| 5h ago | 173 ms | 120 tok/s |
| 5h 15m ago | 120 ms | 163 tok/s |
| 5h 30m ago | 127 ms | 154 tok/s |
| 5h 45m ago | 123 ms | 160 tok/s |
| 6h ago | 121 ms | 157 tok/s |
| 6h 15m ago | 122 ms | 160 tok/s |
| 6h 30m ago | 126 ms | 151 tok/s |
| 6h 45m ago | 124 ms | 160 tok/s |
| 7h ago | 122 ms | 155 tok/s |
| 7h 15m ago | 126 ms | 159 tok/s |
| 7h 30m ago | 117 ms | 163 tok/s |
| 7h 45m ago | 117 ms | 162 tok/s |
| 8h ago | 122 ms | 159 tok/s |
| 8h 15m ago | 118 ms | 163 tok/s |
| 8h 30m ago | 124 ms | 156 tok/s |
| 8h 45m ago | 122 ms | 158 tok/s |
| 9h ago | 119 ms | 159 tok/s |
| 9h 15m ago | 166 ms | 125 tok/s |
| 9h 30m ago | 118 ms | 158 tok/s |
| 9h 45m ago | 118 ms | 158 tok/s |
| 10h ago | 124 ms | 155 tok/s |
| 10h 15m ago | 121 ms | 160 tok/s |
| 10h 30m ago | 113 ms | 165 tok/s |
| 10h 45m ago | 148 ms | 143 tok/s |
| 11h ago | 121 ms | 163 tok/s |
| 11h 15m ago | 122 ms | 163 tok/s |
| 11h 30m ago | 112 ms | 164 tok/s |
| 11h 45m ago | 115 ms | 164 tok/s |
| 12h ago | 113 ms | 166 tok/s |
| 12h 15m ago | 110 ms | 163 tok/s |
| 12h 30m ago | 116 ms | 166 tok/s |
| 12h 45m ago | 121 ms | 160 tok/s |
| 13h ago | 119 ms | 164 tok/s |
| 13h 15m ago | 111 ms | 166 tok/s |
| 13h 30m ago | 111 ms | 167 tok/s |
| 13h 45m ago | 112 ms | 164 tok/s |
| 14h ago | 109 ms | 168 tok/s |
| 14h 15m ago | 110 ms | 164 tok/s |
| 14h 30m ago | 118 ms | 163 tok/s |
| 14h 45m ago | 116 ms | 165 tok/s |
| 15h ago | 120 ms | 159 tok/s |
| 15h 15m ago | 109 ms | 169 tok/s |
| 15h 30m ago | 147 ms | 145 tok/s |
| 15h 45m ago | 115 ms | 160 tok/s |
| 16h ago | 117 ms | 158 tok/s |
| 16h 15m ago | 113 ms | 165 tok/s |
| 16h 30m ago | 110 ms | 162 tok/s |
| 16h 45m ago | 119 ms | 161 tok/s |
| 17h ago | 112 ms | 162 tok/s |
| 17h 15m ago | 120 ms | 161 tok/s |
| 17h 30m ago | 117 ms | 160 tok/s |
| 17h 45m ago | 116 ms | 161 tok/s |
| 18h ago | 121 ms | 162 tok/s |
| 18h 15m ago | 113 ms | 168 tok/s |
| 18h 30m ago | 120 ms | 164 tok/s |
| 18h 45m ago | 121 ms | 163 tok/s |
| 19h ago | 119 ms | 163 tok/s |
| 19h 15m ago | 122 ms | 154 tok/s |
| 19h 30m ago | 122 ms | 160 tok/s |
| 19h 45m ago | 114 ms | 167 tok/s |
| 20h ago | 124 ms | 161 tok/s |
| 20h 15m ago | 113 ms | 164 tok/s |
| 20h 30m ago | 113 ms | 168 tok/s |
| 20h 45m ago | 116 ms | 159 tok/s |
| 21h ago | 121 ms | 163 tok/s |
| 21h 15m ago | 120 ms | 160 tok/s |
| 21h 30m ago | 117 ms | 163 tok/s |
| 21h 45m ago | 115 ms | 167 tok/s |
| 22h ago | 157 ms | 137 tok/s |
| 22h 15m ago | 118 ms | 161 tok/s |
| 22h 30m ago | 124 ms | 157 tok/s |
| 22h 45m ago | 120 ms | 155 tok/s |
| 23h ago | 119 ms | 163 tok/s |
| 23h 15m ago | 121 ms | 159 tok/s |
| 23h 30m ago | 117 ms | 158 tok/s |
| 23h 45m ago | 126 ms | 155 tok/s |
| 24h ago | 127 ms | 155 tok/s |
API
Make a request
The same request shape as every other model on /v1/chat/completions. Switching between them is a change to one string.
curl https://api.weir.example/v1/chat/completions \
-H "Authorization: Bearer $WEIR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "devstral-small-2",
"messages": [
{ "role": "user", "content": "What does a KV cache do?" }
]
}'Dedicated
Need reserved capacity?
Devstral Small 2 runs on the shared serverless pool. Dedicated replicas for it are set up on request, with the same API and the same model ID.
Talk to an engineerCompare