OpenAI: gpt-oss-120b

Text input Text output Free Option
Author's Description

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.

Key Specifications
Cost
$$$
Context
131K
Parameters
120B
Released
Aug 05, 2025
Speed
Ability
Reliability
Supported Parameters

This model supports the following parameters:

Tools Include Reasoning Top Logprobs Tool Choice Max Tokens Reasoning Top P Stop Logprobs Frequency Penalty Temperature Presence Penalty Logit Bias
Features

This model supports the following features:

Reasoning Tools
Performance Summary

OpenAI's gpt-oss-120b, an open-weight 117B-parameter MoE model, demonstrates competitive performance across various benchmarks. It exhibits competitive response times, ranking in the 57th percentile for speed, and offers competitive pricing, placing in the 59th percentile. Notably, the model boasts exceptional reliability with a 97% success rate, indicating consistent and usable responses. The model excels in several key areas. It achieves high accuracy in Coding (93.0%, 90th percentile), standing out as the most accurate model at its price point. General Knowledge (99.5%, 82nd percentile) and Mathematics (94.0%, 91st percentile) also show strong performance. Its Reasoning capabilities are robust (96.0%, 89th percentile), and it demonstrates excellent Email Classification (99.0%, 88th percentile). Furthermore, gpt-oss-120b shows good performance in mitigating hallucinations (98.0% accuracy). However, there are areas for improvement. Instruction Following accuracy is moderate (46.0%, 45th percentile), suggesting potential limitations with complex, multi-layered instructions. Its performance in Keyword Topic Relevance Classification (70.0%, 30th percentile) and Ethics (93.8%, 27th percentile) is comparatively lower, indicating these might be areas for further optimization. Despite these, its overall reliability and strong performance in critical reasoning and knowledge-based tasks make it a promising model for high-reasoning and agentic production use cases.

Model Pricing

Current Pricing

Feature Price (per 1M tokens)
Prompt $0.15
Completion $0.6

Price History

Available Endpoints
Provider Endpoint Name Context Length Pricing (Input) Pricing (Output)
Fireworks
Fireworks | openai/gpt-oss-120b 131K $0.15 / 1M tokens $0.6 / 1M tokens
Groq
Groq | openai/gpt-oss-120b 131K $0.15 / 1M tokens $0.75 / 1M tokens
Cerebras
Cerebras | openai/gpt-oss-120b 131K $0.35 / 1M tokens $0.75 / 1M tokens
BaseTen
BaseTen | openai/gpt-oss-120b 131K $0.1 / 1M tokens $0.5 / 1M tokens
Parasail
Parasail | openai/gpt-oss-120b 131K $0.15 / 1M tokens $0.6 / 1M tokens
Together
Together | openai/gpt-oss-120b 131K $0.15 / 1M tokens $0.6 / 1M tokens
Novita
Novita | openai/gpt-oss-120b 131K $0.1 / 1M tokens $0.5 / 1M tokens
Nebius
Nebius | openai/gpt-oss-120b 131K $0.15 / 1M tokens $0.6 / 1M tokens
DeepInfra
DeepInfra | openai/gpt-oss-120b 131K $0.05 / 1M tokens $0.45 / 1M tokens
AtlasCloud
AtlasCloud | openai/gpt-oss-120b 131K $0.1 / 1M tokens $0.5 / 1M tokens
Chutes
Chutes | openai/gpt-oss-120b 131K $0.05 / 1M tokens $0.25 / 1M tokens
Crusoe
Crusoe | openai/gpt-oss-120b 131K $0.15 / 1M tokens $0.5 / 1M tokens
Phala
Phala | openai/gpt-oss-120b 131K $0.1 / 1M tokens $0.49 / 1M tokens
NCompass
NCompass | openai/gpt-oss-120b 131K $0.072 / 1M tokens $0.28 / 1M tokens
GMICloud
GMICloud | openai/gpt-oss-120b 131K $0.07 / 1M tokens $0.28 / 1M tokens
GMICloud
GMICloud | openai/gpt-oss-120b 131K $0.05 / 1M tokens $0.25 / 1M tokens
WandB
WandB | openai/gpt-oss-120b 131K $0.05 / 1M tokens $0.25 / 1M tokens
SambaNova
SambaNova | openai/gpt-oss-120b 131K $0.22 / 1M tokens $0.59 / 1M tokens
WandB
WandB | openai/gpt-oss-120b 131K $0.15 / 1M tokens $0.6 / 1M tokens
InferenceNet
InferenceNet | openai/gpt-oss-120b 131K $0.05 / 1M tokens $0.45 / 1M tokens
Google
Google | openai/gpt-oss-120b 131K $0.15 / 1M tokens $0.6 / 1M tokens
Benchmark Results
Benchmark Category Reasoning Strategy Free Executions Accuracy Cost Duration
Other Models by openai