gpt-oss-20b: 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 | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
Amazon Bedrockamazon-bedrock | No data | — | — | 99.15% | 99.15% | — | — | — | 131K | $0.070 / $0.150 |
Amazon Bedrockamazon-bedrock/eu-west-1 | No data | — | — | — | — | — | — | — | 131K | $0.070 / $0.150 |
| No data | — | — | 99.24% | 99.24% | — | — | — | 131K | $0.030 / $0.130◆ (best) | |
| No data | — | — | 99.54%◆ (best) | 99.54%◆ (best) | — | — | — | 131K | $0.030 / $0.140◆ (best) | |
Fireworksfireworks | No data | — | — | 97.16% | 97.16% | — | — | — | 131K | $0.070 / $0.300 |
Googlegoogle-vertex/us-central1 | No data | — | — | 79.84% | 79.84% | — | — | — | 131K | $0.070 / $0.250 |
Groqgroq | No data | — | — | 98.17% | 98.17% | — | — | — | 131K | $0.075 / $0.300 |
| No data | — | — | 98.91% | 98.91% | — | — | — | 131K | $0.040 / $0.150 | |
| No data | — | — | 98.75% | 98.75% | — | — | — | 131K | $0.030 / $0.150◆ (best) | |
Phalaphala | No data | — | — | 92.14% | 92.14% | — | — | — | 131K | $0.040 / $0.150 |
| No data | — | — | 97.98% | 97.98% | — | — | — | 131K | $0.040 / $0.180 | |
Togethertogether | No data | — | — | — | — | — | — | — | 131K | $0.050 / $0.200 |
- No data24h —
- 7d
- 99.15%
- 30d
- 99.15%
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 131K
- $/Mtok
- $0.070 / $0.150
amazon-bedrock - No data24h —
- 7d
- —
- 30d
- —
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 131K
- $/Mtok
- $0.070 / $0.150
amazon-bedrock/eu-west-1 - No data24h —
- 7d
- 99.24%
- 30d
- 99.24%
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 131K
- $/Mtok
- $0.030 / $0.130◆ (best)
coreweave/fp4fp4 - No data24h —
- 7d
- 99.54%◆ (best)
- 30d
- 99.54%◆ (best)
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 131K
- $/Mtok
- $0.030 / $0.140◆ (best)
deepinfra/bf16bf16 - No data24h —
- 7d
- 97.16%
- 30d
- 97.16%
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 131K
- $/Mtok
- $0.070 / $0.300
fireworks - No data24h —
- 7d
- 79.84%
- 30d
- 79.84%
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 131K
- $/Mtok
- $0.070 / $0.250
google-vertex/us-central1 - Groq—No data24h —
- 7d
- 98.17%
- 30d
- 98.17%
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 131K
- $/Mtok
- $0.075 / $0.300
groq - No data24h —
- 7d
- 98.91%
- 30d
- 98.91%
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 131K
- $/Mtok
- $0.040 / $0.150
novita/fp4fp4 - No data24h —
- 7d
- 98.75%
- 30d
- 98.75%
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 131K
- $/Mtok
- $0.030 / $0.150◆ (best)
parasail/fp4fp4 - No data24h —
- 7d
- 92.14%
- 30d
- 92.14%
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 131K
- $/Mtok
- $0.040 / $0.150
phala - No data24h —
- 7d
- 97.98%
- 30d
- 97.98%
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 131K
- $/Mtok
- $0.040 / $0.180
siliconflow/fp8fp8 - No data24h —
- 7d
- —
- 30d
- —
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 131K
- $/Mtok
- $0.050 / $0.200
together
No recent measurement. ◆ marks the best value in each column among these endpoints.
Price vs latency
No endpoint currently reports both a price and a latency.
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, averaged over the hour.
Throughput
Median tokens per second.