GPT OSS 20B

OpenAI gpt-oss-20b

Model Information
Slug gpt-oss-20b
Release Date August 5, 2025
Aliases gpt-oss-20b
Organization
Name OpenAI
Description

gpt-oss-20b is an open-weight 21B parameter model released by OpenAI under the Apache 2.0 license. It uses a Mixture-of-Experts (MoE) architecture with 3.6B active parameters per forward pass, optimized for lower-latency inference and deployability on consumer or single-GPU hardware. The model is trained in OpenAI’s Harmony response format and supports reasoning level configuration, fine-tuning, and agentic capabilities including function calling, tool use, and structured outputs. Context: 131072

Available at 15 Providers
Provider Input Price ($/1M) Output Price ($/1M) Free
LMStudio $0.00 $0.00
Chutes $0.02 $0.10
OpenRouter $0.02 $0.10
IO.NET $0.03 $0.14
SiliconFlow (China) $0.04 $0.18
SiliconFlow $0.04 $0.18
Nebius Token Factory $0.05 $0.20
Helicone $0.05 $0.20
FastRouter $0.05 $0.20
OVHcloud AI Endpoints $0.05 $0.18
Fireworks AI $0.05 $0.20
AIHubMix $0.11 $0.55
Cloudflare AI Gateway $0.20 $0.30
Poe $450.00 -
Weights & Biases $5,000.00 $20,000.00