gpt-oss-120b openai/gpt-oss-120b
Served by 30 providers. Cheapest: Requesty at $0.059 blended /1M tokens. The most expensive listing costs 7.6× as much for the same model.
| Provider | Type | Input /1M | Output /1M | Cached input | Blended | Context | Notes | Source |
|---|---|---|---|---|---|---|---|---|
| Requesty runware/gpt-oss-120b |
LLM routers | $0.032 | $0.14 | — | $0.059 | 131k | upstream route runware; geo global; 5 other upstream routes listed (input $0.06-$0.17/M); API list price; Requesty's 5% markup on model cost (pricing page) is not included |
✓ src 2026-09-18 |
| AkashML openai/gpt-oss-120b |
Inference hosts | $0.03 | $0.17 | $0.03 | $0.065 | 128k | Limited Trial. |
✓ src src2 2026-09-18 |
| CoreWeave openai/gpt-oss-120b |
Inference hosts | $0.03 | $0.17 | — | $0.065 | 131k | Context from docs model table (rounded, e.g. "262k"). |
✓ src src2 2026-09-18 |
| DeepInfra openai/gpt-oss-120b |
Inference hosts | $0.037 | $0.17 | — | $0.07 | 131k | bf16 Also sold as openai/gpt-oss-120b-Turbo ($0.15/$0.60) and -Ultra ($0.20/$0.95) speed tiers. |
✓ src src2 2026-09-18 |
| Crusoe | Inference hosts | $0.05 | $0.20 | $0.05 | $0.088 | — | cached tokens price listed equal to input |
✓ src 2026-09-18 |
| GMI Cloud Inference Engine openai/gpt-oss-120b |
Inference hosts | $0.05 | $0.25 | — | $0.10 | — | from GMI console public pricing API (console is client-rendered; docs say prices live in console Model Hub) |
✓ src src2 2026-09-18 |
| Novita AI openai/gpt-oss-120b |
Inference hosts | $0.05 | $0.25 | — | $0.10 | 131k |
✓ src src2 2026-09-18 |
|
| Venice.ai API openai-gpt-oss-120b |
Inference hosts | $0.07 | $0.30 | — | $0.128 | 128k |
✓ src src2 2026-09-18 |
|
| DigitalOcean openai-gpt-oss-120b |
Inference hosts | $0.055 | $0.385 | — | $0.138 | 128k |
✓ src src2 2026-09-18 |
|
| SiliconFlow openai/gpt-oss-120b |
Inference hosts | $0.05 | $0.45 | — | $0.15 | — | context shown on pricing page as "131K" |
✓ src 2026-09-18 |
| Google Vertex AI (Gemini Enterprise Agent Platform) | Hyperscaler AI platforms | $0.09 | $0.36 | — | $0.158 | — | batch −50% Managed MaaS API. Batch $0.045/$0.18. |
✓ src 2026-09-18 |
| Nscale Serverless Inference openai/gpt-oss-120b |
Inference hosts | $0.10 | $0.40 | — | $0.175 | 131k | price as listed by Hugging Face Inference Providers router for provider "nscale" (HF states it passes provider prices through with no markup); Nscale publishes model prices only in its console / authenticated /v1/models |
2° src 2026-09-18 |
| OVHcloud AI Endpoints gpt-oss-120b |
Inference hosts | $0.092 | $0.459 | — | $0.184 | — | fp4 list price EUR 0.08/M in, EUR 0.4/M out; context shown on catalog as "131K" (rounded); converted at ECB EURUSD 1.1481 2026-09-17 (latest ECB reference rate) |
✓ src 2026-09-18 |
| Baseten | Inference hosts | $0.10 | $0.50 | — | $0.20 | — | no cached-input price listed Model API id not shown on pricing page. |
◐ src 2026-09-18 |
| Amazon Bedrock openai.gpt-oss-120b-1:0 |
Hyperscaler AI platforms | $0.15 | $0.60 | — | $0.262 | 128k | batch −50% |
✓ src src2 2026-09-18 |
| Featherless openai/gpt-oss-120b |
Inference hosts | $0.15 | $0.60 | $0.02 | $0.262 | 131k | Developer-plan per-token rate; model_class gpt-oss-120b; also usable flat-rate on $25/mo Chat plan (32K ctx, human chat only) |
✓ src src2 src3 2026-09-18 |
| Fireworks AI accounts/fireworks/models/gpt-oss-120b |
Inference hosts | $0.15 | $0.60 | $0.015 | $0.262 | — | batch −50% Priority $0.18/$0.018/$0.72 |
✓ src 2026-09-18 |
| Groq openai/gpt-oss-120b |
Inference hosts | $0.15 | $0.60 | — | $0.262 | 131k | batch −50% Cached input billed at 50% discount (not listed as a number); ~500 tok/s; Developer plan 250K TPM / 1K RPM |
✓ src 2026-09-18 |
| Microsoft Foundry (Azure AI Foundry / Azure OpenAI) gpt-oss-120b |
Hyperscaler AI platforms | $0.15 | $0.60 | — | $0.262 | — | Global Standard, Azure Retail Prices API meter 'gpt-oss-120B Inp glbl Tokens' (eastus); API meter is per 1K tokens, multiplied by 1000 (unit conversion only); Data Zone $0.165 in / $0.66 out |
✓ src src2 2026-09-18 |
| Nebius | Inference hosts | $0.15 | $0.60 | — | $0.262 | 131k | fp4 Price as passed through by OpenRouter; Nebius price list is login-gated — promote to primary before publication |
2° src 2026-09-18 |
| OpenRouter openai/gpt-oss-120b |
LLM routers | $0.15 | $0.60 | $0.075 | $0.262 | 131k | api modality text->text |
✓ src 2026-09-18 |
| Oracle OCI Generative AI openai.gpt-oss-120b |
Hyperscaler AI platforms | $0.15 | $0.60 | — | $0.262 | — | Parts B112004/B112005. provider_model_id per OCI naming convention not verified on the price list. |
✓ src src2 2026-09-18 |
| Parasail | Inference hosts | $0.10 | $0.75 | $0.055 | $0.263 | — | also "gpt-oss-120b (Fast)" tier $0.15 in/$0.60 out, no cache price |
✓ src 2026-09-18 |
| Together AI openai/gpt-oss-120b |
Inference hosts | $0.15 | $0.60 | — | $0.262 | 131k | mxfp4 |
✓ src 2026-09-18 |
| IBM watsonx.ai gpt-oss-120b |
Hyperscaler AI platforms | $0.159 | $0.636 | — | $0.278 | — |
✓ src src2 2026-09-18 |
|
| Scaleway gpt-oss-120b |
Inference hosts | $0.172 | $0.689 | — | $0.301 | — | batch −50% list price EUR 0.15/M in, EUR 0.6/M out; Paris region; prices before tax; Batches API -50%; converted at ECB EURUSD 1.1481 2026-09-17 (latest ECB reference rate) |
✓ src 2026-09-18 |
| SambaNova Cloud (SambaCloud) gpt-oss-120b |
Inference hosts | $0.22 | $0.59 | — | $0.312 | 128k | Production model. (OpenRouter lists SambaNova at $0.14/$0.95 — differs from SambaNova page) |
✓ src src2 2026-09-18 |
| Replicate openai/gpt-oss-120b |
Inference hosts | $0.18 | $0.72 | — | $0.315 | — |
✓ src 2026-09-18 |
|
| Cerebras Inference gpt-oss-120b |
Inference hosts | $0.35 | $0.75 | — | $0.45 | 131k | Free tier: 65k context, 32k max output; ~3000 tok/s; prompt caching supported (price not listed) |
✓ src 2026-09-18 |
| Cloudflare Workers AI @cf/openai/gpt-oss-120b |
Inference hosts | $0.35 | $0.75 | — | $0.45 | — | Billed in neurons ($0.011/1k neurons); USD per-M-token equivalent as published by Cloudflare |
✓ src 2026-09-18 |
✓ from the provider's own pricing page · ◐ partly confirmed there · 2° third-party source. Router prices may exclude the router's own fees; see each router's page.