PipelineScore
← Back to leaderboard
local

openai/gpt-oss-20b

Released Context 0Kopenai-gpt-oss-20b
PipelineScore
92.0TRUNK
Ranked #1 of 52 models · 98th percentileCode is the headline (100.0); reasoning is the soft spot (75.0). Best-fit profile: Local-first.

Category breakdown

Score per category, normalized 0–100 against the v1 anchor.

Code
100.0
Reason
75.0
Tool Use
87.5
RAG
100.0
Speed
99.3

Strengths

Code100.0
RAG100.0
Speed99.3

Same model, different rigs

Every submission of openai/gpt-oss-20b on the 0–100 scale. The spread is the point: where it runs changes what you get.

0255075100

Best 93.2 on m5-max-48gb · lowest 90.8 on m5-max-48gb · spread 2.4 pts across 2 runs. Hover a dot for its rig.

Sample tasks

A taste of what the test pack measures. Full prompts are private and rotated daily.

CodeDifficulty 1code-fib-1

Fibonacci function

Write a Python `fib(n)` returning the nth Fibonacci number, O(n).

ReasonDifficulty 1reason-math-1

Train meeting time

Two trains, opposite directions, given speeds and start times — when do they meet?

RAGDifficulty 2rag-extract-1

Extract metrics to JSON

From the context, extract net sales, operating margin, and free cash flow as a JSON object. Numbers only.

Tool UseDifficulty 2tool-schema-1

OpenAPI param selection

Given an OpenAPI schema with limit/offset/sort, fill JSON for 'next 50, recent first.'

RAGDifficulty 2rag-grounding-1

Refuses to fabricate

Context lacks the answer — does the model fabricate or correctly say it can't?

Compare with