GPT OSS 120B

OpenAI gpt-oss-120b

Model Information
Slug gpt-oss-120b
Release Date August 5, 2025
Aliases gpt-oss-120b
Organization
Name OpenAI
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. Context: 131072

Available at 21 Providers
Provider Input Price ($/1M) Output Price ($/1M) Free
Nvidia $0.00 $0.00
Cortecs $0.00 $0.00
OpenRouter $0.04 $0.19
Helicone $0.04 $0.16
IO.NET $0.04 $0.40
SiliconFlow (China) $0.05 $0.45
SiliconFlow $0.05 $0.45
OVHcloud AI Endpoints $0.09 $0.47
Synthetic $0.10 $0.10
submodel $0.10 $0.50
Nebius Token Factory $0.15 $0.60
Together AI $0.15 $0.60
FastRouter $0.15 $0.60
Fireworks AI $0.15 $0.60
Scaleway $0.15 $0.60
AIHubMix $0.18 $0.90
Vultr $0.20 $0.20
Cloudflare AI Gateway $0.35 $0.75
NanoGPT $1.00 $2.00
Poe $1,200.00 -
Weights & Biases $15,000.00 $60,000.00