// OpenAI
gpt-oss-120b
A OpenAI model. 12 prices across 12 channels side by side, each with its source and check date – plus specs, licence, speed and benchmarks.
- Official in/out per 1M
- —
- Lowest (DeepInfra)
- $0.037 / $0.17
- Context
- 131K
- AA index
- 11.6
Prices checked 2026-10-22Official sources linked on every rowNo estimatesCheapest LLM APIs →
01Specs at a glance
Official values first; gaps filled from OpenRouter’s API where the vendor page is silent.
| Vendor | OpenAI |
|---|---|
| Context window | 131K (131,072 tokens) |
| Max output | 118K |
| Input | Text |
| Output | Text |
| Tool calling / JSON | Yes / Yes |
| Weights & licence | Open weights · Apache-2.0 · Hugging Face |
| Parameters | 117B (from weight files) |
| Released | Aug 5, 2025 |
| Model IDs | openai/gpt-oss-120b · gpt-oss-120b · openai-gpt-oss-120b |
02Prices on 12 channels
Per 1M tokens, in the currency each source publishes (1 USD = 6.7 CNY for the blended column). Sorted: official first, then cheapest.
| Channel | Source | Input | Output | Cache hit | Blended $ | Context | Notes |
|---|---|---|---|---|---|---|---|
| DeepInfra · bf16 Via OpenRouter / provider listed there | OpenRouter 10-22 | $0.037 | $0.17 | — | 0.07 | 131K | DeepInfra price as listed on OpenRouter (pass-through; quantization=bf16; discount field=0); not checked against DeepInfra's own page |
| Crusoe · bf16 Via OpenRouter / provider listed there | OpenRouter 10-22 | $0.05 | $0.25 | $0.05 | 0.1 | 131K | Crusoe price as listed on OpenRouter (pass-through; quantization=bf16; discount field=0); not checked against Crusoe's own page |
| Novita · fp4 Via OpenRouter / provider listed there | OpenRouter 10-22 | $0.05 | $0.25 | — | 0.1 | 131K | Novita price as listed on OpenRouter (pass-through; quantization=fp4; discount field=0); not checked against Novita's own page |
| BaseTen · fp4 Via OpenRouter / provider listed there | OpenRouter 10-22 | $0.1 | $0.5 | $0.1 | 0.2 | 128K | BaseTen price as listed on OpenRouter (pass-through; quantization=fp4; discount field=0); not checked against BaseTen's own page |
| Together Inference platform (own price page) | Platform 10-10 | $0.15 | $0.6 | — | 0.263 | 131K | Together AI own pricing page |
| Groq Inference platform (own price page) | Platform 10-10 | $0.15 | $0.6 | — | 0.263 | 131K | Groq own pricing page; Groq-reported speed ~500 tok/s |
| Fireworks Inference platform (own price page) | Platform 10-10 | $0.15 | $0.6 | $0.015 | 0.263 | 131K | Batch −50%Fireworks AI own pricing page; batch -50% |
| Parasail · fp4 Via OpenRouter / provider listed there | OpenRouter 10-22 | $0.1 | $0.75 | $0.055 | 0.263 | 131K | Parasail price as listed on OpenRouter (pass-through; quantization=fp4; discount field=0); not checked against Parasail's own page |
| Nebius · fp4 Via OpenRouter / provider listed there | OpenRouter 10-22 | $0.15 | $0.6 | — | 0.263 | 131K | Nebius price as listed on OpenRouter (pass-through; quantization=fp4; discount field=0); not checked against Nebius's own page |
| SiliconFlow · fp8 Via OpenRouter / provider listed there | OpenRouter 10-22 | $0.15 | $0.6 | $0.075 | 0.263 | 131K | SiliconFlow price as listed on OpenRouter (pass-through; quantization=fp8; discount field=0); not checked against SiliconFlow's own page |
| SambaNova · unknown Via OpenRouter / provider listed there | OpenRouter 10-22 | $0.14 | $0.95 | — | 0.343 | 131K | SambaNova price as listed on OpenRouter (pass-through; quantization=unknown; discount field=0); not checked against SambaNova's own page |
| Cerebras · fp16 Via OpenRouter / provider listed there | OpenRouter 10-22 | $0.35 | $0.75 | $0.35 | 0.45 | 131K | Cerebras price as listed on OpenRouter (pass-through; quantization=fp16; discount field=0); not checked against Cerebras's own page |
“Official” = vendor pricing page; “Platform” = the platform’s own pricing page; “OpenRouter” = price listed on OpenRouter, not checked on the platform’s site. Click a source tag to open the original page.
03Where you can use it
Derived only from the price sources above. Country restrictions are set by each vendor – check their terms before you sign up.
- Via OpenRouter / provider listed there
DeepInfra · Crusoe · Novita · BaseTen · Parasail · Nebius · SiliconFlow · SambaNova · Cerebras - Inference platform (own price page)
Together · Groq · Fireworks - Self-host the open weights
GPU / VRAM sizing and build plans→ - Free option
gpt-oss-120b→ - Free option
gpt-oss-120b (Groq)→ - Free option
gpt-oss-120b (Cerebras trial)→
04Performance
All results →Speed and AA index from Artificial Analysis public leaderboards (not our own tests); benchmark scores from our benchmark library, best result per benchmark.
- AA indexgpt-oss-120b (High)
- 11.6
- Output speed (median)
- 233 tok/s
- Time to first tokengpt-oss-120b (Low): 0.85s
- 0.83s
- AA index per $ (blended)
- 165.17
| Benchmark | Best score | Source | Date |
|---|---|---|---|
| LMArena Text | 1352 | official | 2026-10-08 |
| GPQA Diamond | 75.8% | independent | 2025-12-11 |
| MMLU-Pro | 80.8% | vendor | 2026-03-11 |
| SWE-bench Verified | 26.0% | official | 2025-08-07 |
| AIME (OTIS Mock AIME 2024–2025) | 88.9% | independent | 2025-12-11 |
| Terminal-Bench 2.0 | 18.7% | official | 2025-11-01 |
05Price changes we recorded
All price changes →From our daily re-checks of the sources above.
No price change recorded since we started tracking (2026-10-10).
06More from OpenAI
FAQ
How much does the gpt-oss-120b API cost?
There is no official pay-as-you-go price in our data. The lowest listed price is $0.037 / $0.17 on DeepInfra.
What is the context window of gpt-oss-120b?
131K tokens of context, up to 118K output tokens. Platforms may cap it lower – see the table.
Is gpt-oss-120b open source?
Yes, the weights are published on Hugging Face (openai/gpt-oss-120b). Licence: Apache-2.0.