OpenAI: gpt-oss-20b
Analysis Summary
OpenAI's gpt-oss-20b is a compact open-weight model with a strong benchmark profile for its size and price. A livecodebench score of 0.777 and agentic index of 35.4 place it ahead of many larger models on coding and tool-use tasks. Instruction following (ifbench 0.65) is particularly strong, and the math index of 89.3 is best-in-class for this price tier.
For businesses, gpt-oss-20b is well-suited to high-volume coding assistance, structured output generation, and lightweight agentic pipelines where cost efficiency is critical. Long-context reasoning (lcr 0.31) is moderate, so very complex multi-step document analysis may benefit from a larger model.
At $0.029 input and $0.14 output per million tokens, this is one of the most cost-effective verified models available. Teams running large-scale automation or wanting a reliable coding assistant at minimal cost should consider it a primary option.
Assessed June 30, 2026
Editorial notes
OpenAI gpt-oss-20b pairs a livecodebench score of 0.777, agentic index of 35.4, and strong instruction following at $0.029 input, making it an exceptional value proposition.
Rankings consider pricing, capabilities, benchmarks, and real-world applicability and are refreshed as new models launch. Feedback?
Performance Profile
How OpenAI: gpt-oss-20b compares
OpenAI: gpt-oss-20b ranks #177 of 385 AI models we track for overall intelligence, #81 of 129 for coding, #129 of 293 for agentic tasks. Its 131K-token context window is larger than 59% of the models we list. At $0.03 per million input tokens it is cheaper than 76% of comparable models.
About OpenAI: gpt-oss-20b
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..
Capabilities
Performance Indices
Source: Artificial Analysis
Benchmark Scores
Intelligence
Technical
Content
Benchmark data from Artificial Analysis and Hugging Face
How does OpenAI: gpt-oss-20b stack up?
Compare side-by-side with other professional models.
Model Information
| OpenRouter ID |
openai/gpt-oss-20b
|
| Provider | openai |
| Release Date | August 5, 2025 |
| Context Length | 131,072 tokens |
| Status | Active |
Pricing
| Token Type | Cost per 1M tokens | Cost per 1K tokens |
|---|---|---|
| Input | $0.03 | $0.000029 |
| Output | $0.14 | $0.000140 |
Live Performance
Live endpoint metrics, refreshed every 30 minutes.
Leaderboard Categories
External Resources
Frequently asked questions about OpenAI: gpt-oss-20b
How much does OpenAI: gpt-oss-20b cost?
OpenAI: gpt-oss-20b costs $0.03 per million input tokens and $0.14 per million output tokens.
What is the context window of OpenAI: gpt-oss-20b?
OpenAI: gpt-oss-20b has a context window of 131,072 tokens (131K).
Is OpenAI: gpt-oss-20b good for coding?
On our coding benchmark index, OpenAI: gpt-oss-20b ranks #81 of 129 models, placing it in the broader range of the field for code generation and debugging.
What can OpenAI: gpt-oss-20b do?
OpenAI: gpt-oss-20b supports tool use and function calling.
Who created OpenAI: gpt-oss-20b?
OpenAI: gpt-oss-20b is developed by OpenAI and was released on August 5, 2025.
