Model profile / OpenAI

GPT-OSS 120B

OpenAI's self-hosted 117B-parameter open-weight model with configurable reasoning, a 131,072-token context and maximum output, released under Apache 2.0.

5 Aug 2025Open weightsProvider

Sources & details

Four areas of evidence

The capabilities behind the model.

How the indices work

Reasoning

No current rating

Mathematical and scientific problem solving.

Awaiting a qualified category score

Coding

No current rating

Software engineering and programming.

Awaiting a qualified category score

General

No current rating

Factual reliability and instruction following.

Awaiting a qualified category score

Agentic

85.8Index points

Tool use, workflows and professional tasks.

This model│ Other models · 40 rated

Score and research range

Full-precision point: 85.7561503221697
Range: 84.33–87.16 · 90% Research Range
Scope: Agentic score under joint reference/source reweighting conditional on admitted W/X/P native outcomes

Score uses the admitted direct native evidence under the selected method.

11retained native records

9evidence families

Capabilities estimateQualified original evidence across four areas.

Current index snapshot · 2026-09-16. Each category uses its own scale; scores across categories are not directly comparable. Evidence counts describe the retained Capabilities inputs; a family can contain several tasks or configurations. A point lead does not establish superiority.

Performance in context

Capabilities, price & speed.

Explore every model

GPT-OSS 120B is highlighted wherever a compatible measurement is available.

GPT-OSS 120B: OpenAI open weights; no first-party hosted API token price

Lumina Capabilities Index

Capabilities versus API price

The current Capabilities cohort, with this model highlighted.

Lumina

2025+ · 62 models · 1 highlighted

Chart loads as you explore

Lumina Capabilities Index

Capabilities versus output speed

The current Capabilities cohort, with this model highlighted.

Lumina

2025+ · 72 models · Developer API · 10K · 1 highlighted

Chart loads as you explore

Explore the detail

Sources & model details.

Original evidence, with its context.

Current index evidence 2Scores, research ranges and the sources behind each category.

Capabilities

Score uses the admitted direct native evidence under the selected method.

Exact checkpoint: gpt-oss-120b
Method: methods-a4471cee84fd478d2ed3
Range interpretation: C1 native factor under conditional crossed source/family reweighting of admitted fixed-corpus reference and own portfolio
Sources: Artificial Analysis:Artificial Analysis · Epoch AI:Epoch AI · Zapier:Zapier
Source cutoff: {"asOfMeaning":"Selected local candidate: retained historical measurements plus reviewed additions; execution dates remain source-native.","newIdentityRecovery":"2026-09-13T08:41:08.804270+00:00","publicProjectionDate":"2026-09-16T06:02:45.491398+00:00","scoring":"Inherited pinned baseline source vintages plus retained4A/4A.1 originals and SWE owner correction; heterogeneous execution dates, often unknown. Not a claim of simultaneous2026-09-13 execution."}

Explore this index

Agentic

Score uses the admitted direct native evidence under the selected method.

Exact checkpoint: gpt-oss-120b
Method: methods-a4471cee84fd478d2ed3
Range interpretation: Agentic score under joint reference/source reweighting conditional on admitted W/X/P native outcomes
Sources: Artificial Analysis · Zapier
Source cutoff: {"asOfMeaning":"Selected local candidate: retained historical measurements plus reviewed additions; execution dates remain source-native.","newIdentityRecovery":"2026-09-13T08:41:08.804270+00:00","publicProjectionDate":"2026-09-16T06:02:45.491398+00:00","scoring":"Inherited pinned baseline source vintages plus retained4A/4A.1 originals and SWE owner correction; heterogeneous execution dates, often unknown. Not a claim of simultaneous2026-09-13 execution."}

Explore this index

Benchmark measurements 77Published values in each benchmark’s own units.

77 measurements

GPT-OSS 120B published benchmark measurements
Benchmark	Result	Published configuration	Source
Artificial Analysis Agentic Index2026	13.17 index	Exact BenchLM registry variant GPT-OSS 120B; bulk export does not retain a complete upstream harness configuration.Source-native system	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21source-checked · reference-only
Artificial Analysis Agentic Index2026	13.17 index	Exact BenchLM registry variant GPT-OSS 120B; bulk export does not retain a complete upstream harness configuration.Source-native system	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27source-checked · reference-only
Artificial Analysis Agentic Index2026	13.17 index	Exact BenchLM registry variant GPT-OSS 120B; bulk export does not retain a complete upstream harness configuration.Source-native system	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01source-checked · reference-only
Artificial Analysis AIME 20252026	93.4%	Exact BenchLM registry variant GPT-OSS 120B; bulk export does not retain a complete upstream harness configuration.Source-native system	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21source-checked · reference-only
Artificial Analysis AIME 20252026	93.4%	Exact BenchLM registry variant GPT-OSS 120B; bulk export does not retain a complete upstream harness configuration.Source-native system	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27source-checked · reference-only
Artificial Analysis AIME 20252026	93.4%	Exact BenchLM registry variant GPT-OSS 120B; bulk export does not retain a complete upstream harness configuration.Source-native system	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01source-checked · reference-only
Artificial Analysis Coding Index2026	30.44 index	Exact BenchLM registry variant GPT-OSS 120B; bulk export does not retain a complete upstream harness configuration.Source-native system	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21source-checked · reference-only
Artificial Analysis Coding Index2026	30.44 index	Exact BenchLM registry variant GPT-OSS 120B; bulk export does not retain a complete upstream harness configuration.Source-native system	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27source-checked · reference-only
Artificial Analysis Coding Index2026	30.44 index	Exact BenchLM registry variant GPT-OSS 120B; bulk export does not retain a complete upstream harness configuration.Source-native system	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01source-checked · reference-only
Artificial Analysis EnterpriseOps-Gym2026	25.5%	Exact BenchLM registry variant GPT-OSS 120B; bulk export does not retain a complete upstream harness configuration.Source-native system	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21source-checked · reference-only

Showing 10 of 77 measurements

Specifications & source recordsDocumented model details and original references.

Model ID / checkpoint: gpt-oss-120b
Weight access: Open weights
Context window: 131,072
Maximum output: 131,072
Modalities: text
Details checked: 20 Aug 2026

Specialist task evidenceSeparate agent runs and execution conditions.

Specialist evidence

Agent runs and task economics

Source-native operational evidence for GPT-OSS 120B. Exact configurations remain separate. Costs are comparable only within the same selected benchmark/evaluation, and these rows never enter Overall Score.

Specialist only · ranking prohibited

Evaluation	Exact configuration	Performance	Cost / task	Tokens / task	Execution
SWE-bench Owner Leaderboards · owner-current · bash-only SWE-bench Owner Leaderboards · checked 2026-08-29	gpt-oss-120b-high gpt-oss-120b	26.0%	$0.057	—	27.6 calls
SWE-bench Owner Leaderboards · owner-current · verified SWE-bench Owner Leaderboards · checked 2026-08-29	gpt-oss-120b-high mini-SWE-agent + gpt-oss-120b	26.0%	$0.057	—	27.6 calls

All models

Model profile / OpenAI

GPT-OSS 120B

OpenAI's self-hosted 117B-parameter open-weight model with configurable reasoning, a 131,072-token context and maximum output, released under Apache 2.0.

5 Aug 2025Open weightsProvider

Sources & details

Four areas of evidence

The capabilities behind the model.

How the indices work

Reasoning

No current rating

Mathematical and scientific problem solving.

Awaiting a qualified category score

Coding

No current rating

Software engineering and programming.

Awaiting a qualified category score

General

No current rating

Factual reliability and instruction following.

Awaiting a qualified category score

Agentic

85.8Index points

Tool use, workflows and professional tasks.

This model│ Other models · 40 rated

Score and research range

Full-precision point: 85.7561503221697
Range: 84.33–87.16 · 90% Research Range
Scope: Agentic score under joint reference/source reweighting conditional on admitted W/X/P native outcomes

Score uses the admitted direct native evidence under the selected method.

11retained native records

9evidence families

Capabilities estimateQualified original evidence across four areas.

Performance in context

Capabilities, price & speed.

Explore every model

GPT-OSS 120B is highlighted wherever a compatible measurement is available.

GPT-OSS 120B: OpenAI open weights; no first-party hosted API token price

Lumina Capabilities Index

Capabilities versus API price

The current Capabilities cohort, with this model highlighted.

Lumina

2025+ · 62 models · 1 highlighted

Chart loads as you explore

Lumina Capabilities Index

Capabilities versus output speed

The current Capabilities cohort, with this model highlighted.

Lumina

2025+ · 72 models · Developer API · 10K · 1 highlighted

Chart loads as you explore

Explore the detail

Sources & model details.

Original evidence, with its context.

Current index evidence 2Scores, research ranges and the sources behind each category.

Capabilities

Score uses the admitted direct native evidence under the selected method.

Exact checkpoint: gpt-oss-120b
Method: methods-a4471cee84fd478d2ed3
Range interpretation: C1 native factor under conditional crossed source/family reweighting of admitted fixed-corpus reference and own portfolio
Sources: Artificial Analysis:Artificial Analysis · Epoch AI:Epoch AI · Zapier:Zapier
Source cutoff: {"asOfMeaning":"Selected local candidate: retained historical measurements plus reviewed additions; execution dates remain source-native.","newIdentityRecovery":"2026-09-13T08:41:08.804270+00:00","publicProjectionDate":"2026-09-16T06:02:45.491398+00:00","scoring":"Inherited pinned baseline source vintages plus retained4A/4A.1 originals and SWE owner correction; heterogeneous execution dates, often unknown. Not a claim of simultaneous2026-09-13 execution."}

Explore this index

Agentic

Score uses the admitted direct native evidence under the selected method.

Exact checkpoint: gpt-oss-120b
Method: methods-a4471cee84fd478d2ed3
Range interpretation: Agentic score under joint reference/source reweighting conditional on admitted W/X/P native outcomes
Sources: Artificial Analysis · Zapier
Source cutoff: {"asOfMeaning":"Selected local candidate: retained historical measurements plus reviewed additions; execution dates remain source-native.","newIdentityRecovery":"2026-09-13T08:41:08.804270+00:00","publicProjectionDate":"2026-09-16T06:02:45.491398+00:00","scoring":"Inherited pinned baseline source vintages plus retained4A/4A.1 originals and SWE owner correction; heterogeneous execution dates, often unknown. Not a claim of simultaneous2026-09-13 execution."}

Explore this index

Benchmark measurements 77Published values in each benchmark’s own units.

77 measurements

GPT-OSS 120B published benchmark measurements
Benchmark	Result	Published configuration	Source
Artificial Analysis Agentic Index2026	13.17 index	Exact BenchLM registry variant GPT-OSS 120B; bulk export does not retain a complete upstream harness configuration.Source-native system	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21source-checked · reference-only
Artificial Analysis Agentic Index2026	13.17 index	Exact BenchLM registry variant GPT-OSS 120B; bulk export does not retain a complete upstream harness configuration.Source-native system	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27source-checked · reference-only
Artificial Analysis Agentic Index2026	13.17 index	Exact BenchLM registry variant GPT-OSS 120B; bulk export does not retain a complete upstream harness configuration.Source-native system	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01source-checked · reference-only
Artificial Analysis AIME 20252026	93.4%	Exact BenchLM registry variant GPT-OSS 120B; bulk export does not retain a complete upstream harness configuration.Source-native system	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21source-checked · reference-only
Artificial Analysis AIME 20252026	93.4%	Exact BenchLM registry variant GPT-OSS 120B; bulk export does not retain a complete upstream harness configuration.Source-native system	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27source-checked · reference-only
Artificial Analysis AIME 20252026	93.4%	Exact BenchLM registry variant GPT-OSS 120B; bulk export does not retain a complete upstream harness configuration.Source-native system	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01source-checked · reference-only
Artificial Analysis Coding Index2026	30.44 index	Exact BenchLM registry variant GPT-OSS 120B; bulk export does not retain a complete upstream harness configuration.Source-native system	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21source-checked · reference-only
Artificial Analysis Coding Index2026	30.44 index	Exact BenchLM registry variant GPT-OSS 120B; bulk export does not retain a complete upstream harness configuration.Source-native system	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27source-checked · reference-only
Artificial Analysis Coding Index2026	30.44 index	Exact BenchLM registry variant GPT-OSS 120B; bulk export does not retain a complete upstream harness configuration.Source-native system	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01source-checked · reference-only
Artificial Analysis EnterpriseOps-Gym2026	25.5%	Exact BenchLM registry variant GPT-OSS 120B; bulk export does not retain a complete upstream harness configuration.Source-native system	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21source-checked · reference-only

Showing 10 of 77 measurements

Specifications & source recordsDocumented model details and original references.

Model ID / checkpoint: gpt-oss-120b
Weight access: Open weights
Context window: 131,072
Maximum output: 131,072
Modalities: text
Details checked: 20 Aug 2026

Specialist task evidenceSeparate agent runs and execution conditions.

Specialist evidence

Agent runs and task economics

Specialist only · ranking prohibited

Evaluation	Exact configuration	Performance	Cost / task	Tokens / task	Execution
SWE-bench Owner Leaderboards · owner-current · bash-only SWE-bench Owner Leaderboards · checked 2026-08-29	gpt-oss-120b-high gpt-oss-120b	26.0%	$0.057	—	27.6 calls
SWE-bench Owner Leaderboards · owner-current · verified SWE-bench Owner Leaderboards · checked 2026-08-29	gpt-oss-120b-high mini-SWE-agent + gpt-oss-120b	26.0%	$0.057	—	27.6 calls

The capabilities behind the model.

Capabilities, price & speed.

Capabilities versus API price

Capabilities versus output speed

Sources & model details.

Capabilities

Agentic

Agent runs and task economics

More from OpenAI.

GPT-6 Astra

GPT-5.6 Sol

GPT-5.6 Terra

The capabilities behind the model.

Capabilities, price & speed.

Capabilities versus API price

Capabilities versus output speed

Sources & model details.

Capabilities

Agentic

Agent runs and task economics

More from OpenAI.

GPT-6 Astra

GPT-5.6 Sol

GPT-5.6 Terra