gpt-oss-120b API pricing
OpenAI · Budget price tier · Apache 2.0 · released 2025-08. Prices in USD per 1 million tokens. Data snapshot: 2026-10-22.
gpt-oss-120b has held the same list price since 2026-10-18 and is the 85th cheapest of 164 models in the Budget price tier. Buying from Near AI rather than at the price we quote lowers the token price by about 75%, for a quantized fp4 build. It is rated a stable release but does not meet all four criteria of the FinOps Friendly badge.
- Input: $0.15 per 1M tokens
- Output: $0.60 per 1M tokens
- Batch API (in/out): $0.0296 / $0.136 per 1M tokens
- Prompt-cache read: $0.075 per 1M tokens
- Context window: 131K tokens
- Arena ELO: 1352 as of 2026-10-08
Available on
AWS Bedrock regions: ap-northeast-1, eu-central-1, eu-west-1, us-east-1, us-east-2, us-west-2
Market position
On blended price ($0.465 per 1M tokens at 30% input, 70% output) it is cheaper than 44% of models in the Budget price tier: 85th cheapest of 164, tied with 7 others. Across every price tier it is cheaper than 70% of priced text models in the catalogue: 85th cheapest of 308, tied with 7 others. Its Arena ELO of 1352 ranks 122nd of the 156 models that carry one, tied with one other, above 21% of them. With an Arena score of 1352, gpt-oss-120b sits below OptimNow's lowest working quality threshold, Standard (Arena 1415+), so the quality-bar method would not shortlist it for any of the 8 use cases the table prices. Each threshold is a distance below the best ranked model, which scores 1525 on Arena scores as of 8 Oct 2026.
Licence, size and capabilities
gpt-oss-120b is open source: its weights are published under Apache 2.0, an OSI-approved licence with no field-of-use limit, so it can be self-hosted as well as bought by the token. Its published size is 120B parameters. Listed capabilities: Text, Reasoning, Agents. Filling its 131K context window once costs $0.0197 in input tokens at list price.
Pricing by host
A host is a company that runs the model and sells it by the token: the model maker, a cloud platform like Amazon Bedrock, Azure or Vertex, or an independent host. One model, several hosts, and often several prices. This model is sold by 22 different hosts. The price we quote elsewhere on this page is marked. We quote one host's rate so that models stay comparable; it is not a recommendation to buy there.
Cheapest host: Near AI at $0.03 input / $0.15 output per 1M tokens, 75% below the price we quote on the blended price (30% input, 70% output). It serves a quantized fp4 build, which is not the same product as the full-precision model. The dearest, SambaNova, charges 6.2x the cheapest.
| Host | Input $/1M | Output $/1M | Precision |
|---|---|---|---|
| Near AI | $0.03 | $0.15 | fp4 |
| Venice | $0.03 | $0.15 | int4 |
| CoreWeave | $0.03 | $0.17 | fp4 |
| DeepInfra | $0.037 | $0.17 | bf16 |
| DekaLLM | $0.03 | $0.18 | bf16 |
| AkashML | $0.03 | $0.187 | bf16 |
| Crusoe | $0.05 | $0.25 | bf16 |
| Mancer 2 | $0.05 | $0.25 | fp8 |
| Novita | $0.05 | $0.25 | fp4 |
| $0.09 | $0.36 | not declared | |
| DigitalOcean | $0.06 | $0.42 | not declared |
| BaseTen | $0.10 | $0.50 | fp4 |
| Amazon Bedrock (price we quote) | $0.15 | $0.60 | not declared |
| Nebius (price we quote) | $0.15 | $0.60 | fp4 |
| SiliconFlow (price we quote) | $0.15 | $0.60 | fp8 |
| Phala (price we quote) | $0.15 | $0.60 | not declared |
| Together (price we quote) | $0.15 | $0.60 | not declared |
| Groq (price we quote) | $0.15 | $0.60 | not declared |
| Parasail | $0.10 | $0.75 | fp4 |
| Mara | $0.15 | $0.75 | not declared |
| Cerebras | $0.35 | $0.75 | fp16 |
| SambaNova | $0.14 | $0.95 | not declared |
List prices as published on 2026-10-22, via OpenRouter. Negotiated rates, committed-use discounts and provisioned capacity are not shown. These hosts do not all serve the model at the same numeric precision, so the cheapest is not a like-for-like substitute for the dearest.
Monthly cost at 100K requests, by business use case
From the after-discounts figure (prompt caching on the cached input share, batch API for async workloads, only where OpenAI publishes those rates), which assumes you implement both, to the list price. One figure means no discount is published.
| Use case | Tokens (in/out) | Monthly (after discounts – list) |
|---|---|---|
| Support Ticket | 1,500 / 500 | $46 – $53 |
| Knowledge Q&A | 2,000 / 800 | $68 – $78 |
| Meeting Summary | 10,000 / 1,200 | $46 – $222 |
| Marketing Content | 2,500 / 1,800 | $142 – $146 |
| Coding Task | 3,000 / 2,000 | $154 – $165 |
| Invoice Processing | 1,500 / 600 | $13 – $59 |
| Call Summary | 2,000 / 700 | $15 – $72 |
| Agent Workflow | 6,000 / 3,000 | $239 – $270 |
Price history
gpt-oss-120b entered our daily price archive on 2026-10-18; tracking began on 2026-08-15. It listed at $0.15 input / $0.60 output per 1M tokens that day. No list price change since 2026-10-18. Last checked against the 2026-10-22 snapshot.
Alternatives to gpt-oss-120b
- Next cheaper model in the Budget price tier: Hunyuan A13B Instruct (Tencent) at $0.441 blended, 5% less.
- Next more expensive: Saba (Mistral) at $0.48 blended, 3% more.
The nearest models in the Budget price tier on Arena ELO, each with its blended price and score difference against gpt-oss-120b.
- GLM 4.5V (Zhipu): $0.60 in / $1.80 out per 1M, Arena ELO 1352 (210% more expensive, same Arena ELO)
- Nemotron 3.5 Lightning (Nvidia): $0.07 in / $0.20 out per 1M, Arena ELO 1349 (65% cheaper, 3 ELO points lower)
- Mistral Small 3.2 24B (Mistral): $0.0938 in / $0.25 out per 1M, Arena ELO 1356 (56% cheaper, 4 ELO points higher)
- Qwen3 32B (Alibaba): $0.08 in / $0.28 out per 1M, Arena ELO 1347 (53% cheaper, 5 ELO points lower)
- Cheapest within 15 Arena points: Gemma 3 12B (Google), 74% cheaper, 10 ELO points lower, out of 17 models in that band.
- Highest Arena ELO within 20% of this price: MiMo-V2.5-Pro (Xiaomi), 11% more expensive, 116 ELO points higher, out of 14 models in that price band.
Other models from OpenAI
The 8 nearest to gpt-oss-120b in price, of 53 other OpenAI models in the catalogue. Prices per 1M tokens, input then output.
- GPT-4.1 Nano: $0.10 / $0.40
- GPT-6 Luna: $0.10 / $0.50
- GPT-6 Luna Pro: $0.10 / $0.50
- GPT-4o-mini: $0.15 / $0.60
- GPT-5.6 Luna: $0.20 / $1.20
- GPT-5.6 Luna Pro: $0.20 / $1.20
- GPT-5.4 Nano: $0.20 / $1.25
- GPT-4.1 Mini: $0.40 / $1.60
All 54 OpenAI models and prices
Common questions about gpt-oss-120b pricing
- How much does gpt-oss-120b cost?
- gpt-oss-120b costs $0.15 per 1M input tokens and $0.60 per 1M output tokens at Amazon Bedrock, the host whose rate we quote. Prices as listed on 2026-10-22.
- Which host is cheapest for gpt-oss-120b?
- Cheapest host: Near AI at $0.03 input / $0.15 output per 1M tokens, 75% below the price we quote on the blended price (30% input, 70% output). It serves a quantized fp4 build, which is not the same product as the full-precision model. The dearest, SambaNova, charges 6.2x the cheapest. Host rates as listed on 2026-10-22.
- Does gpt-oss-120b offer batch or cached pricing?
- Yes, both. Batch API pricing is $0.0296 per 1M input tokens and $0.136 per 1M output tokens, and prompt-cache reads cost $0.075 per 1M tokens, as listed on 2026-10-22.
- What would gpt-oss-120b cost per month?
- With the Support Ticket profile (1,500 input and 500 output tokens per request), 100,000 requests a month cost about $53 on gpt-oss-120b at the list price quoted on 2026-10-22.
- Can gpt-oss-120b be self-hosted?
- Yes. gpt-oss-120b is open source: its weights are published under Apache 2.0, an OSI-approved licence with no field-of-use limit, so it can be self-hosted as well as bought by the token.
- Has the price of gpt-oss-120b changed?
- gpt-oss-120b entered our daily price archive on 2026-10-18; tracking began on 2026-08-15. It listed at $0.15 input / $0.60 output per 1M tokens that day. No list price change since 2026-10-18. Last checked against the 2026-10-22 snapshot.
These are list prices. Your real cost also depends on caching and batching, negotiated discounts, marketplace agreements, and the contracts you already have. OptimNow audits exactly that.
Build the business case · Get a FinOps review
LLM pricing · Models by maker · Cloud compute pricing · Compare models · Price barometer · FinOps guides · API and MCP server