gpt-oss-20b

openai/gpt-oss-20b

OpenAI’s gpt-oss-20b is a text‑only model released on 2024‑06‑01. It supports a context window of 131,072 tokens, includes tool usage and reasoning capabilities, and is priced at $0.03 per million input tokens and $0.13 per million output tokens. The model’s weights are not publicly available.

  • tool use
  • structured outputs
  • reasoning
  • open-weight

At a glance

provideropenai
releasedJan 1, 1970
context131K
input / M$0.030

Specifications

What the API gives you

Context window
131K tokens
Max output
118K tokens
Input price
$0.030 / 1M tokens
Output price
$0.13 / 1M tokens
Cache read
$0.030 / 1M tokens
Modalities
text → text
Knowledge cutoff
Jun 30, 2024
Released
Jan 1, 1970

Measured evidence

Public benchmark record

Design Arena Website
877 Elo2026-07-28
Design Arena Website
877 Elo2026-07-30
Design Arena Website
877 Elo2026-07-27
Design Arena Website
877 Elo2026-07-29
Design Arena Website
877 Elo2026-07-31
Design Arena Website
876 Elo2026-08-02
Design Arena Website
876 Elo2026-08-01
Design Arena Website
876 Elo2026-08-05
Design Arena Website
876 Elo2026-08-03
Design Arena Website
875 Elo2026-08-09
Design Arena Website
876 Elo2026-08-04
Design Arena Website
872 Elo2026-08-14
Design Arena Website
876 Elo2026-08-06
Design Arena Website
876 Elo2026-08-07
Design Arena Website
875 Elo2026-08-10
Design Arena Website
875 Elo2026-08-08
Design Arena Website
864 Elo2026-08-18
Design Arena Website
875 Elo2026-08-11
Design Arena Website
864 Elo2026-08-15
Design Arena Website
874 Elo2026-08-12
Design Arena Website
873 Elo2026-08-13
Design Arena Website
864 Elo2026-08-19
Design Arena Website
863 Elo2026-08-16
Design Arena Website
864 Elo2026-08-20
Design Arena Website
872 Elo2026-08-27
Design Arena Website
864 Elo2026-08-17
Design Arena Website
872 Elo2026-08-21
Design Arena Website
872 Elo2026-08-22
Design Arena Website
872 Elo2026-08-23
Design Arena Website
873 Elo2026-08-24
Design Arena Website
872 Elo2026-08-28
Design Arena Website
873 Elo2026-08-25
Design Arena Website
873 Elo2026-08-26
Design Arena Website
872 Elo2026-08-29
Design Arena Website
872 Elo2026-08-30
Design Arena Website
872 Elo2026-08-31
Design Arena Website
872 Elo2026-09-01
Design Arena Website
871 Elo2026-09-02
Design Arena Website
869 Elo2026-09-03
Design Arena Website
868 Elo2026-09-04
LMArena Creative Writing
1199 Elo2026-07-21
LMArena Creative Writing
1199 Elo2026-07-27
LMArena Creative Writing
1199 Elo2026-07-30
LMArena Creative Writing
1199 Elo2026-08-02
LMArena Creative Writing
1199 Elo2026-08-03
LMArena Creative Writing
1199 Elo2026-08-06
LMArena Creative Writing
1199 Elo2026-08-12
LMArena Creative Writing
1198 Elo2026-08-21
LMArena Creative Writing
1201 Elo2026-08-27
LMArena Creative Writing
1201 Elo2026-09-01
LMArena Long Query
1250 Elo2026-07-21
LMArena Long Query
1250 Elo2026-07-27
LMArena Long Query
1250 Elo2026-07-30
LMArena Long Query
1250 Elo2026-08-02
LMArena Long Query
1250 Elo2026-08-03
LMArena Long Query
1251 Elo2026-08-06
LMArena Long Query
1251 Elo2026-08-12
LMArena Long Query
1251 Elo2026-08-21
LMArena Long Query
1251 Elo2026-08-27
LMArena Long Query
1251 Elo2026-09-01
LMArena Math
1318 Elo2026-07-21
LMArena Math
1318 Elo2026-07-27
LMArena Math
1318 Elo2026-07-30
LMArena Math
1318 Elo2026-08-02
LMArena Math
1318 Elo2026-08-03
LMArena Math
1317 Elo2026-08-06
LMArena Math
1317 Elo2026-08-12
LMArena Math
1317 Elo2026-08-21
LMArena Math
1317 Elo2026-08-27
LMArena Math
1317 Elo2026-09-01
LMArena Text
1317 Elo2026-07-21
LMArena Text
1317 Elo2026-07-27
LMArena Text
1317 Elo2026-07-30
LMArena Text
1317 Elo2026-08-02
LMArena Text
1318 Elo2026-08-10
LMArena Text
1317 Elo2026-08-06
LMArena Text
1317 Elo2026-08-03
LMArena Text
1318 Elo2026-08-21
LMArena Text
1318 Elo2026-08-12
LMArena Text
1317 Elo2026-08-27
LMArena Text
1317 Elo2026-09-01
SWE-rebench
8.1% ± 1.12025-09-01tools

LMArena data licensed CC-BY-4.0. Aider (Apache-2.0) and SWE-bench (MIT) are open. Design Arena uses blind human preference. Submitted scaffold plus model result. Compare the harness and effort settings, not only the model name.

Open-weight model on Hugging Face.

Task evidence

How broad is the support?

Each status is exact to this model version and the selected task. Official release claims remain provisional.

  • Best LLM for Business WritingProvisional evidence

    70% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following. 5 of 5 configured evaluation feeds currently have results.

  • Best LLM for Marketing CopyProvisional evidence

    70% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following. 5 of 5 configured evaluation feeds currently have results.

  • Best LLM for Chat & RoleplayProvisional evidence

    70% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or LiveBench Language. 5 of 5 configured evaluation feeds currently have results.

  • 70% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or LiveBench Language. 5 of 5 configured evaluation feeds currently have results.

  • Best LLM for Creative WritingProvisional evidence

    60% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or LiveBench Language. 5 of 5 configured evaluation feeds currently have results.

  • Best LLM for WritingProvisional evidence

    60% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or LiveBench Language. 5 of 5 configured evaluation feeds currently have results.

  • 55% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or LiveBench Reasoning. 5 of 5 configured evaluation feeds currently have results.

  • 50% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or LiveBench Reasoning. 5 of 5 configured evaluation feeds currently have results.

  • 40% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from Aider Polyglot or LiveBench Coding. 6 of 6 configured evaluation feeds currently have results.

  • Best LLM for Competition MathProvisional evidence

    40% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Mathematics or LiveBench Reasoning. 5 of 5 configured evaluation feeds currently have results.

  • Best LLM for Math & ReasoningProvisional evidence

    40% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Mathematics or LiveBench Reasoning. 5 of 5 configured evaluation feeds currently have results.

  • Best Value LLMProvisional evidence

    36% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or Route reliability. 3 of 4 configured evaluation feeds currently have results.

  • 33% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or Route reliability. 3 of 4 configured evaluation feeds currently have results.

  • Best LLM for Autonomous AgentsProvisional evidence

    29% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from Frontier-Bench or LiveBench Agentic Coding. 7 of 7 configured evaluation feeds currently have results.

  • Best LLM for Fast SummariesProvisional evidence

    29% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or Route reliability. 3 of 4 configured evaluation feeds currently have results.

  • Best LLM for CodingProvisional evidence

    26% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from Aider Polyglot or Design Arena Coding. 8 of 8 configured evaluation feeds currently have results.

  • 25% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or LiveBench Language. 4 of 5 configured evaluation feeds currently have results.

  • Best Cheap LLM for High VolumeProvisional evidence

    25% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or Route reliability. 3 of 4 configured evaluation feeds currently have results.

  • 25% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or LiveBench Language. 4 of 5 configured evaluation feeds currently have results.

  • Cheapest Capable LLMProvisional evidence

    25% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or Route reliability. 3 of 4 configured evaluation feeds currently have results.

  • Best LLM for TranslationProvisional evidence

    25% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or LiveBench Language. 4 of 5 configured evaluation feeds currently have results.

  • Best LLM for Massive ContextProvisional evidence

    25% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LMArena Document or LongBench v2. 3 of 4 configured evaluation feeds currently have results.

  • Best LLM for Agentic CodingProvisional evidence

    23% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from Design Arena Full-stack or Design Arena Web-apps. 8 of 8 configured evaluation feeds currently have results.

  • 20% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from FACTS Grounding or LiveBench Instruction Following. 4 of 6 configured evaluation feeds currently have results.

  • Best LLM for SummarizationProvisional evidence

    20% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from FACTS Grounding or LiveBench Instruction Following. 4 of 6 configured evaluation feeds currently have results.

  • Best Local LLM for CodingProvisional evidence

    20% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from Aider Polyglot or LiveBench Coding. 5 of 5 configured evaluation feeds currently have results.

  • Best LLM for AI AgentsProvisional evidence

    20% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from Berkeley Function Calling or LiveBench Agentic Coding. 7 of 7 configured evaluation feeds currently have results.

  • Best LLM for Hermes AgentProvisional evidence

    18% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from Berkeley Function Calling or LMArena Agent. 5 of 5 configured evaluation feeds currently have results.

  • Best LLM for OpenHandsProvisional evidence

    18% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from Berkeley Function Calling or LMArena Agent. 5 of 5 configured evaluation feeds currently have results.

  • Best LLM for GooseProvisional evidence

    18% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from Berkeley Function Calling or LMArena Agent. 5 of 5 configured evaluation feeds currently have results.

  • 18% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from FACTS Grounding or LMArena Document. 4 of 6 configured evaluation feeds currently have results.

  • Best LLM for OpenClawProvisional evidence

    18% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from Berkeley Function Calling or LMArena Agent. 5 of 5 configured evaluation feeds currently have results.

  • 17% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from Berkeley Function Calling or LiveBench Data Analysis. 4 of 5 configured evaluation feeds currently have results.

  • 17% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from Berkeley Function Calling or LiveBench Data Analysis. 4 of 5 configured evaluation feeds currently have results.

  • Best LLM for Frontend & UIProvisional evidence

    15% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from Design Arena Coding or Design Arena UI. 5 of 5 configured evaluation feeds currently have results.

  • 13% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LiveBench Data Analysis or Route reliability. 2 of 4 configured evaluation feeds currently have results.

  • Best LLM for Code CompletionProvisional evidence

    13% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from Aider Polyglot or LiveBench Coding. 4 of 4 configured evaluation feeds currently have results.

  • 10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from Berkeley Function Calling or LMArena Agent. 4 of 5 configured evaluation feeds currently have results.

  • 10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from Berkeley Function Calling or LMArena Agent. 4 of 4 configured evaluation feeds currently have results.

  • Best LLM for OCR & DocumentsProvisional evidence

    10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LMArena Document or LMArena Vision. 3 of 5 configured evaluation feeds currently have results.

  • 10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LiveBench Reasoning or LMArena Vision. 3 of 4 configured evaluation feeds currently have results.

  • 10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LMArena Document or LMArena Vision. 3 of 5 configured evaluation feeds currently have results.

  • Best LLM for Grounded RAGProvisional evidence

    10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from CRAG or FACTS Grounding. 2 of 5 configured evaluation feeds currently have results.

  • Best LLM for Charts & DiagramsProvisional evidence

    10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LMArena Document or LMArena Vision. 3 of 5 configured evaluation feeds currently have results.

  • Best LLM for Document ParsingProvisional evidence

    10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LMArena Document or LMArena Vision. 3 of 5 configured evaluation feeds currently have results.

Recommendation coverage

Where this model appears

Not currently included in a published recommendation. The task evidence panel above says exactly what is confirmed and what is still missing.

Catalog data via OpenRouter and models.dev. Last checked within the past 6 hours.