Mistral Small 3.2 24B: providers compared
Every endpoint serving this model, side by side. ◆ marks the best value in each column — per model, never fleet-wide.
Current state
| State | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Healthy | 100%◆ (best) | 99.80% | 99.82% | 99.82% | 679ms | 4.84s | 32 tps | 128K | $0.075 / $0.200◆ (best) | |
| Healthy | 100%◆ (best) | 99.97% | 99.88% | 99.82% | 302ms◆ (best) | 2.23s◆ (best) | 55 tps◆ (best) | 131K | $0.090 / $0.300 | |
| Healthy | 100%◆ (best) | 99.76% | 99.54% | 99.23% | 854ms | 6.00s | 31 tps | 256K | $0.094 / $0.250 | |
Mistralmistral/eu | No data | — | 100% | 99.99%◆ (best) | 100%◆ (best) | 1.88s | 7.15s | 42 tps | 33K | $0.100 / $0.300 |
Mistralmistral/zdr | No data | — | — | — | — | — | — | — | 131K | $0.110 / $0.330 |
- DeepInfra100%◆ (best)Healthy24h 99.80%
- 7d
- 99.82%
- 30d
- 99.82%
- p50
- 679ms
- p99
- 4.84s
- Throughput
- 32 tps
- Context
- 128K
- $/Mtok
- $0.075 / $0.200◆ (best)
deepinfra/fp8fp8 - Parasail100%◆ (best)Healthy24h 99.97%
- 7d
- 99.88%
- 30d
- 99.82%
- p50
- 302ms◆ (best)
- p99
- 2.23s◆ (best)
- Throughput
- 55 tps◆ (best)
- Context
- 131K
- $/Mtok
- $0.090 / $0.300
parasail/bf16bf16 - Venice100%◆ (best)Healthy24h 99.76%
- 7d
- 99.54%
- 30d
- 99.23%
- p50
- 854ms
- p99
- 6.00s
- Throughput
- 31 tps
- Context
- 256K
- $/Mtok
- $0.094 / $0.250
venice/fp8fp8 - No data24h 100%
- 7d
- 99.99%◆ (best)
- 30d
- 100%◆ (best)
- p50
- 1.88s
- p99
- 7.15s
- Throughput
- 42 tps
- Context
- 33K
- $/Mtok
- $0.100 / $0.300
mistral/eu - No data24h —
- 7d
- —
- 30d
- —
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 131K
- $/Mtok
- $0.110 / $0.330
mistral/zdr
Measured 1m ago. Latency and throughput are OpenRouter's rolling 30-minute windows. ◆ marks the best value in each column among these endpoints.
Price vs latency
One dot per endpoint at the latest round · further left = faster, lower = cheaper · hover a dot for its tag
History
Availability
Axis starts at 90% so small differences stay visible; it expands to 0 when an endpoint falls below.
Time to first token
Logarithmic axis — the fleet spans two orders of magnitude. P50 is OpenRouter's rolling 30-minute window.
Throughput
Median tokens per second.