Long documents
Declared, not measured
Context window: 1M tokens as declared. It says how much fits, not how well the model uses it.
$0.21
mixed / 1M tokens
Mixed price per 1M tokens: 3 input tokens to 1 output token.
in $0.12 · out $0.48 · Cached input $0.0024
via River
Price disputed
Models that think before they answer can use many more output tokens, so your real cost can be higher than the figure shown.
US dollars per million tokens, converted to your currency at the NBP rate.
Each line names the measurement it rests on. A task that is not listed has no measurement or declared feature we can show.
Declared, not measured
Context window: 1M tokens as declared. It says how much fits, not how well the model uses it.
Declared, not measured
Accepts: images. Declared by the provider through OpenRouter; we did not test it.
The price used in the catalog is the vendor's own; when the vendor has no API of its own, it is the cheapest regular host.
| Provider | Input | Output | Cached input | Mixed | Notes |
|---|---|---|---|---|---|
| Baidu | $0.0507 | $0.2028 | $0.001 | $0.0887 58% cheaper | fp8 · promo −83% |
| Wafer | $0.051 | $0.567 | $0.018 | $0.18 14% cheaper | US · mxfp4 · Numeric precision used by this host. Lower precision (fp4, int4) is cheaper and can change quality. |
| River Price used here | $0.12 | $0.48 | $0.0024 | $0.21 | |
| DeepInfra | $0.14 | $0.42 | $0.0042 | $0.21 | fp8 · promo −30% |
| Relace | $0.12 | $0.50 | $0.003 | $0.215 | fp4 · Numeric precision used by this host. Lower precision (fp4, int4) is cheaper and can change quality. |
| Inference.net | $0.103 | $0.60 | $0.008 | $0.2273 | US · fp8 |
| Inference.net | $0.103 | $0.60 | $0.008 | $0.2273 | fp8 |
| AtlasCloud | $0.135 | $0.54 | $0.0135 | $0.2363 | fp8 |
| OpenInference | $0.06 | $0.80 | $0.002 | $0.245 | fp4 · Numeric precision used by this host. Lower precision (fp4, int4) is cheaper and can change quality. |
| Reka | $0.12 | $0.64 | $0.008 | $0.25 | promo −20% |
| StreamLake | $0.147 | $0.588 | $0.0029 | $0.2573 | fp8 · promo −51% |
| CoreWeave | $0.13 | $0.65 | $0.002 | $0.26 | fp8 |
| Novita | $0.15 | $0.60 | $0.003 | $0.2625 | fp8 · promo −50% |
| Cloudflare | $0.10 | $0.90 | $0.03 | $0.30 | |
| Wafer | $0.19 | $0.65 | $0.012 | $0.305 | mxfp4 · Numeric precision used by this host. Lower precision (fp4, int4) is cheaper and can change quality. |
| Cohere | $0.195 | $0.66 | $0.003 | $0.3113 | fp8 |
| Alibaba | $0.18 | $0.72 | $0.0036 | $0.315 | time-of-day price |
| GMICloud | $0.18 | $0.72 | $0.0036 | $0.315 | fp8 · promo −40% |
| Ionstream | $0.20 | $0.70 | $0.02 | $0.325 | |
| DigitalOcean | $0.195 | $0.78 | $0.0039 | $0.3413 | |
| Mancer 2 | $0.2125 | $0.75 | $0.07 | $0.3469 | fp8 |
| Makora | $0.26 | $1.17 | $0.0058 | $0.4875 | fp8 |
| Crusoe | $0.29 | $1.20 | $0.007 | $0.5175 | fp8 |
| BaseTen | $0.30 | $1.20 | $0.007 | $0.525 | US |
| BaseTen | $0.30 | $1.20 | $0.007 | $0.525 | |
| Fireworks | $0.30 | $1.20 | $0.006 | $0.525 | |
| $0.30 | $1.20 | $0.03 | $0.525 | ||
| Modal | $0.30 | $1.20 | $0.03 | $0.525 | |
| Nebius | $0.30 | $1.20 | $0.01 | $0.525 | fp8 |
| Parasail | $0.30 | $1.20 | $0.006 | $0.525 | fp8 |
| SiliconFlow | $0.30 | $1.20 | $0.006 | $0.525 | fp8 |
| Together | $0.30 | $1.20 | $0.006 | $0.525 | |
| Venice | $0.30 | $1.20 | $0.0075 | $0.525 | fp8 |
| Fireworks | $0.45 | $1.80 | $0.009 | $0.7875 | US |
As listed by OpenRouter. "Not listed" means OpenRouter does not list it for this model.
No benchmark score is available for this model.
Prices: OpenRouter routes and a second price table, checked November 21, 2026. Specifications: OpenRouter, November 21, 2026. Exchange rate: NBP, November 20, 2026. Benchmarks: Artificial Analysis and Design Arena via OpenRouter.