OpenAI: gpt-oss-20b (free)

OpenAI: gpt-oss-20b (free)

openai · Released Aug 5, 2025 Efficient
Intelligence #225 / 544
37.2 Our Score
Speed #10 / 257
311.8 tokens / sec
Input
Not priced
Output
Not priced
Context #207 / 551
131,072 tokens

Analysis Summary

OpenAI: gpt-oss-20b (free) sits in the Efficient tier on our leaderboard, ranked #225 of 544 published models on overall intelligence. At $0.000 input and $0.000 output per 1M tokens, it is among the most expensive on the market. It offers a standard large context window and supports tool use, function calling, and reasoning.

Editorial notes

GPT-OSS-20B (free) from OpenAI is available at no cost with a 24.5 intelligence index, strong livecodebench (0.777), and tool use support; an excellent value entry point for lighter business tasks despite modest coding and agentic scores.

Assessed May 5, 2026

Rankings consider pricing, capabilities, benchmarks, and real-world applicability and are refreshed as new models launch. Feedback?

Performance Profile

Intelligence4.7Technical4.5Value0Content5
Intelligence 4.7/10
Technical 4.5/10
Content 5/10
Value 0/10

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..

20B Parameters

Capabilities

Tool Use Function Calling

Performance Indices

Source: Artificial Analysis

24.5 Intelligence Index
18.5 Coding Index
89.3 Math Index

Benchmark Scores

Intelligence

GPQA Diamond 68.8% Graduate-level scientific reasoning
HLE 9.8% Humanity's Last Exam
MMLU Pro 74.8% Multi-task language understanding
SciCode 34.4% Scientific computing

Technical

LiveCodeBench 77.7% Live coding evaluation

Benchmark data from Artificial Analysis and Hugging Face

How does OpenAI: gpt-oss-20b (free) stack up?

Compare side-by-side with other efficient models.

Compare Models

Model Information

OpenRouter ID openai/gpt-oss-20b:free
Provideropenai
Release Date August 5, 2025
Context Length131,072 tokens
Max Completion8,192 tokens
Status Active

Live Performance

Live endpoint metrics — refreshed every 30 minutes.

95.5%
Avg Uptime
9,083ms
Best Latency (TTFT)
12 tok/s
Best Throughput
1/1
Active Endpoints
Available via: OpenInference

Leaderboard Categories