Ways to buy 12
| Way to buy | Input | Output | In / out per 1M | Cached | Vs cheapest |
|---|---|---|---|---|---|
| Darkbloom fp8 | $0.018 | $0.09 | $0.018 / $0.09Cheapest | n/a | Cheapest |
| AkashML fp4 | $0.02 | $0.1 | $0.02 / $0.1+11% vs cheapest | n/a | +11% vs cheapest |
 CoreWeavefp4 · cached $0.03 | $0.03 | $0.13 | $0.03 / $0.13+53% vs cheapest | $0.03 | +53% vs cheapest |
| DekaLLM bf16 | $0.029 | $0.14 | $0.029 / $0.14+58% vs cheapest | n/a | +58% vs cheapest |
 DeepInfrabf16 | $0.03 | $0.14 | $0.03 / $0.14+60% vs cheapest | n/a | +60% vs cheapest |
 Parasailfp4 · cached $0.02 | $0.03 | $0.15 | $0.03 / $0.15+67% vs cheapest | $0.02 | +67% vs cheapest |
 Novita AIfp4 | $0.04 | $0.15 | $0.04 / $0.15+88% vs cheapest | n/a | +88% vs cheapest |
 SiliconFlowfp8 | $0.04 | $0.18 | $0.04 / $0.18+108% vs cheapest | n/a | +108% vs cheapest |
 AWS BedrockDefault region | $0.07 | $0.15 | $0.07 / $0.15+150% vs cheapest | n/a | +150% vs cheapest |
| AWS Bedrock eu-west-1 | $0.07 | $0.15 | $0.07 / $0.15+150% vs cheapest | n/a | +150% vs cheapest |
 Google Vertex AIus-central1 | $0.07 | $0.25 | $0.07 / $0.25+219% vs cheapest | n/a | +219% vs cheapest |
 GroqDefault region · cached $0.037 | $0.075 | $0.3 | $0.075 / $0.3+265% vs cheapest | $0.037 | +265% vs cheapest |
USD per 1M tokens. Regions are separate routes.
Through a gateway
Your own numbers100M input and 20M output tokens a month: $3.60 direct.
Helicone$3.60+0%
Kong AI Gateway$3.60+0%
LiteLLM$3.60+0%
Vercel AI Gateway$3.60+0%
Cloudflare AI Gateway$3.78+5%
Requesty$3.78+5%
Eden AI$3.80+5.5%
OpenRouter$3.80+5.5%
No published token fee: Martian Gateway, Portkey, TrueFoundry AI Gateway.
Related models
| In / out per 1M | Cheapest route | ||||||
|---|---|---|---|---|---|---|---|
 GPT-6 LunaOpenAI · 1.05MCheapest: $0.1 on OpenAI (flex) | $0.1 | $0.5 | $0.1 / $0.5 | $0.2 | 1.05M | $0.1 on OpenAI (flex) | · |
| GPT-6 Luna Pro OpenAI · 1.05MCheapest: $0.1 on OpenAI (flex) | $0.1 | $0.5 | $0.1 / $0.5 | $0.2 | 1.05M | $0.1 on OpenAI (flex) | · |
| GPT-6 Sol OpenAI · 1.05MCheapest: $2 on OpenAI (flex) | $2 | $10 | $2 / $10 | $4 | 1.05M | $2 on OpenAI (flex) | · |
| GPT-6 Sol Pro OpenAI · 1.05MCheapest: $2 on OpenAI (flex) | $2 | $10 | $2 / $10 | $4 | 1.05M | $2 on OpenAI (flex) | · |
 GLM 5.3 FlashZ.ai · 1.31M | $0.045 | $0.14 | $0.045 / $0.14 | $0.069 | 1.31M | on InferenceNet (fp4) | · |
 DeepSeek V4 Flash 0731DeepSeek · 1.31M | $0.053 | $0.158 | $0.053 / $0.158 | $0.079 | 1.31M | on StreamLake (fp8) | · |
 Qwen3.7 FlashQwen (Alibaba) · 1M | $0.03 | $0.13 | $0.03 / $0.13 | $0.055 | 1M | Lab list price | · |
| DeepSeek V4 Flash 0423 DeepSeek · 1M | $0.049 | $0.098 | $0.049 / $0.098 | $0.061 | 1M | on Baidu (fp8) | · |
About gpt-oss-20b
gpt-oss-20b is an open-weight 21B parameter model released by OpenAI under the Apache 2.0 license.
Takes text in and returns text. First listed Sep 23, 2026. Up to 33K output tokens. Blended price $0.036 per 1M tokens at 3 input to 1 output. Model id gpt-oss-20b.
Price history
No list-price history: this model is sold by hosts rather than its lab.
Questions
How much does gpt-oss-20b cost?
$0.018 per million input tokens and $0.09 per million output tokens.
What is the context window of gpt-oss-20b?
131K tokens, with up to 33K tokens of output.
Where is gpt-oss-20b cheapest?
Of 12 routes tracked, Darkbloom is cheapest at $0.018 input and $0.09 output per million tokens.
Sources: OpenRouter's public model catalog and each provider's own pricing page, read every morning at 05:00 UTC; gateway fees re-read every Monday. List prices: your contract may differ. Logos via logo.dev; trademarks belong to their owners.