BenchmarkActivecategory · codingfamily · swe-bench · variant full

SWE-bench (full test split)

swebench.com

resolve real GitHub issues (2,294 instances)

data quality57

Updated 6 h ago · first seen 11 Sept 2026

Metric: resolved · % ↑
Current results: 21
Models: 14
Current leader: gpt-4o 14.6%

Filters

Comparability group

Trust level

Views

Score history · gpt-4o 3 rows

gpt-4o

11.99%date=2024-07-28 · board=Test · system=SWE-agent · model_tag=gpt-4o-2024-05-1328 Jul 2024
18.83%date=2024-06-28 · board=Test · system=AutoCodeRover · model_tag=gpt-4o-2024-05-1328 Jun 2024
14.6%date=2024-06-15 · board=Test · system=AppMap Navie · model_tag=gpt-4o-2024-05-1315 Jun 2024

Back to the leaderboard

Frontier over time · resolved · board=Test · system=AppMap Navie

1 leader change recorded, all dated 15 Jun 2024 — the frontier line needs at least two distinct dates. The corpus is young: every result was first observed on the same day, so leader changes will separate in time as sources are re-crawled.

14.6%gpt-4o OpenAI Official board15 Jun 2024

Includes closed rows (history). A point is emitted whenever a result beats every earlier result of the same group, ordered by evaluated_at when the source publishes it, else observed_at.

Leaderboard 1 models

Select models with +, then Compare.

Leaderboard
#	Model	Score	Trust	Configuration	vs leader	Evaluated	Source	Actions
1	gpt-4oClosedOpenAI · GPT 4	14.6%	Official board	group defaults	leader	15 Jun 2024	swebench.comT2	History

1 results

One row per canonical model — its best current row inside this comparability group (effort variants are folded into the model). Bars are relative to the page's best score. “vs leader” reads comparability: partially comparable = same task, conditions differ (reasoning effort, temperature, judge). Rules →

Definition

Task: Source:AI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2observed 16 h agomedium
Metric: Source:AI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2observed 16 h agomedium
Metric range: Source:AI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2observed 6 h agomedium
Variant: Source:AI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2observed 6 h agomedium
Harness: Source:AI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2observed 6 h agomedium
Comparability note: Source:AI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2observed 6 h agomedium
Paper: Source:AI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2observed 16 h agomedium
Website: swebench.com
Source:AI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2observed 16 h agomedium

Active group

resolved · board=Test · system=AppMap Navie

1 models · 1 current rows

Comparability

Comparable same task and conditions · Partially comparable same task, conditions differ (effort, temperature, judge) · Not comparable different variant or metric

Also known as

SWE-bench (full test split), SWE-bench Full, SWE-bench Test

slug swe-bench-full · bench_01M293SPG8K6FQ2ZZD4YS1NK9R

Rows are grouped by comparability group = canonical metric × config_key (hash of the task-defining configuration keys: variant, evaluator, harness, shots, pass regime…). A leaderboard shows one row per canonical model: its best current row inside the group. Reasoning effort, temperature or judge differences keep rows in the same group but mark them partially comparable. delta_rank compares with the ranking built from the closed (previous) rows of the same group.