Question 1

What is GPT OSS 120B?

Accepted Answer

GPT OSS 120B is a model from OpenAI, released on Aug 5, 2025. gpt-oss-120b is an open-weight, 117B-parameter Mixture-of-Experts (MoE) language model from OpenAI designed for high-reasoning, agentic, and general-purpose production use cases. It activates 5.1B parameters per forward pass and is optimized to run on a single H100 GPU with native MXFP4 quantization. The model supports configurable reasoning depth, full chain-of-thought access, and native tool use, including function calling, browsing, and structured output generation.

Question 2

How much does GPT OSS 120B cost?

Accepted Answer

On the Respan gateway, input starts at $0.04 per 1M tokens on DeepInfra and OpenRouter, and output starts at $0.17 per 1M tokens on DeepInfra and OpenRouter, after gateway discounts. Averaged across 20 providers, it costs $0.14 per 1M input tokens and $0.54 per 1M output tokens.

Question 3

Which providers serve GPT OSS 120B on Respan?

Accepted Answer

20 providers serve GPT OSS 120B: DeepInfra, OpenRouter, Novita, Tera, Google Vertex AI, Baseten, and 9 more. Tera, Google Vertex AI, Fireworks, Groq, and 1 more require your own provider API key (BYOK).

Question 4

What is GPT OSS 120B's context window?

Accepted Answer

GPT OSS 120B accepts up to 131K tokens of context on its largest provider route.

Question 5

How do I use GPT OSS 120B with the Respan gateway?

Accepted Answer

Point any OpenAI-compatible SDK at https://api.respan.ai/api/ with your Respan API key and set the model to "gpt-oss-120b". To set it up with a coding agent, copy the agent setup prompt at the top of this page.

Question 6

How fast is GPT OSS 120B?

Accepted Answer

Over the last 30 days of requests through the Respan gateway, the fastest provider was Nebius at 115 tokens per second, and the lowest response latency was 0.5 seconds on Parasail. Speed varies with request size and provider.

Question 7

Where does GPT OSS 120B rank on the Respan Index?

Accepted Answer

GPT OSS 120B scores 14.5 out of 100 on the Respan Index v1.1, ranking #124 of 138 models. By category: Coding 7.4, Agents 7.9, Reasoning 14.0, Knowledge & language 12.4 (estimated), and Human preference 39.8. The Respan Index combines independent and lab-reported benchmark results into one score, using benchmark data collected by llmmetric.com.

Question 8

How does GPT OSS 120B score on benchmarks?

Accepted Answer

GPT OSS 120B has 12 published benchmark scores, including SWE-bench Verified 62.4%, APEX-Agents 4.4%, and GPQA Diamond 80.1%. Each score links to where it was published; the data is collected by llmmetric.com.

Provider	Context	Input	Output
DeepInfra`deepinfra/openai/gpt-oss-120b`	131K	$0.04/M	$0.17/M
OpenRouter`openrouter/openai/gpt-oss-120b`	131K	$0.04/M	$0.17/M
NovitaSave 10%`novita/openai/gpt-oss-120b`	131K	$0.05/M	$0.23/M
TeraBYOK only`tera/openai/gpt-oss-120b`	131K	$0.09/M	$0.36/M
Google Vertex AIBYOK only`vertex_ai/openai/gpt-oss-120b-maas`	131K	$0.09/M	$0.36/M
Baseten`baseten/openai/gpt-oss-120b`	128K	$0.10/M	$0.50/M
Parasail`parasail/openai/gpt-oss-120b`	131K	$0.10/M	$0.75/M
Bedrock`bedrock/openai.gpt-oss-120b-1:0`	131K	$0.15/M	$0.60/M
DeepInfra`deepinfra/openai/gpt-oss-120b-Turbo`	131K	$0.15/M	$0.60/M
Featherless`featherless_ai/openai/gpt-oss-120b`	131K	$0.15/M	$0.60/M
FireworksBYOK only`fireworks_ai/gpt-oss-120b`	131K	$0.15/M	$0.60/M
GroqBYOK only`groq/openai/gpt-oss-120b`	131K	$0.15/M	$0.60/M
Nebius`nebius/openai/gpt-oss-120b`	131K	$0.15/M	$0.60/M
Parasail`parasail/openai/gpt-oss-120b-fast`	131K	$0.15/M	$0.60/M
Parasail`parasail/parasail-gpt-oss-120b`	131K	$0.15/M	$0.60/M
Parasail`parasail/parasail-gpt-oss-120b-fast`	131K	$0.15/M	$0.60/M
TogetherAIBYOK only`together_ai/openai/gpt-oss-120b`	131K	$0.15/M	$0.60/M
IO.net`ionet/openai/gpt-oss-120b`	131K	$0.17/M	$0.63/M
DeepInfra`deepinfra/openai/gpt-oss-120b-Ultra`	131K	$0.20/M	$0.95/M
Cerebras`cerebras/gpt-oss-120b`	131K	$0.35/M	$0.75/M

GPT OSS 120B

Providers

Certified where it counts.

Pricing

Cost by volume

Throughput and latency

Output speed

Time to first token

Response latency

Benchmarks

Respan Indexv1.1

Coding2

Agents

Apps

GPT OSS 120B FAQ

Works with your entire stack

Built for AI agents.
Break less.
Ship more.

Reasoning5

Math2

Arena1