EGI-BENCH
Agents work across eight enterprise evaluations. The suite combines their results into six scores covering activities such as planning, retrieving evidence, and using tools.
- Made by
- Salesforce AI Research
- Released
- October 2026
- Kind
- Simulated companies
- Level
- University
- Scored
- Checked automatically
- Answers
- Agent task
- Tasks
- All public
- Measures
- Capability
- Language
- English
- Interaction
- Agentic
- Environment
- APIs and tools
- Refresh
- One fixed set
- Run by
- Anyone
- In views
- Agents, Law and finance
- Strongest model tested
- GPT-5.5 #21 on the capability index
- Last model added
- April 2026
- Scores refreshed
- November 20th, 2026
GPT-5.5 leads with 65.6 among models released by April 2026 GPT-5.5 leads with 65.6 among models released by April 2026
Best score per entry from the benchmark's paper, read on November 20th, 2026. A paper’s table doesn’t add newer models. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. This view shows only models we track, each at its best.
The newest model with a score here came out in April 2026, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:
- GPT-6.2 Astra
- Claude Fable 5.2
-
Gemini 4 Carbon
- GPT-6 Astra
- and 16 more
- 65.6 GPT-5.5
- 65.3 Claude Sonnet 4.5
- 63.5
Gemini 3.1 Pro
- 63.3 Claude Opus 4.5
- 59.4 GPT-4o (snapshot unspecified)
- 59.1 Claude Haiku 4.5
- 59.1 Gemini 2.5 Pro (snapshot unspecified)
- 58.3 GPT-4.1
- 51.9 GPT-5 mini
- 51.4
Gemini 2.5 Flash
- 50.8 Claude 3.5 Haiku
- 49.3
Gemini 2.0 Flash
- 45.1 GPT-5 nano
- 39.6 GPT-4o mini
- 1 GPT-5.5 quick 65.6
- 2 Claude Sonnet 4.5 quick 65.3
- 3
Gemini 3.1 Pro quick 63.5
- 4 Claude Opus 4.5 quick 63.3
- 5 GPT-4o (snapshot unspecified) quick 59.4
- 6 Claude Haiku 4.5 quick 59.1
- 7 Gemini 2.5 Pro (snapshot unspecified) quick 59.1
- 8 GPT-4.1 quick 58.3
- 9 GPT-5 mini quick 51.9
- 10
Gemini 2.5 Flash Sep 2025, quick 51.4
- 11 Claude 3.5 Haiku quick 50.8
- 12
Gemini 2.0 Flash Feb 2025, quick 49.3
- 13 GPT-5 nano quick 45.1
- 14 GPT-4o mini quick 39.6
Source: The EGI-BENCH paper
- 1 GPT-5.5 quick 65.6
- 2 Claude Sonnet 4.5 quick 65.3
- 3
Gemini 3.1 Pro quick 63.5
- 4 Claude Opus 4.5 quick 63.3
- 5 Claude Haiku 4.5 quick 59.1
- 6 GPT-4.1 quick 58.3
- 7 GPT-5 mini quick 51.9
- 8
Gemini 2.5 Flash Sep 2025, quick 51.4
- 9 Claude 3.5 Haiku quick 50.8
- 10
Gemini 2.0 Flash Feb 2025, quick 49.3
- 11 GPT-5 nano quick 45.1
- 12 GPT-4o mini quick 39.6
Source: The EGI-BENCH paper
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | GPT-5.5 | 65.6 | quick | Apr 2026 | Official leaderboard |
| 2 | Claude Sonnet 4.5 | 65.3 | quick | Sep 2025 | Official leaderboard |
| 3 | | 63.5 | quick | Feb 2026 | Official leaderboard |
| 4 | Claude Opus 4.5 | 63.3 | quick | Nov 2025 | Official leaderboard |
| 5 | GPT-4o (snapshot unspecified) | 59.4 | quick | Official leaderboard | |
| 6 | Claude Haiku 4.5 | 59.1 | quick | Oct 2025 | Official leaderboard |
| 7 | Gemini 2.5 Pro (snapshot unspecified) | 59.1 | quick | Official leaderboard | |
| 8 | GPT-4.1 | 58.3 | quick | Apr 2025 | Official leaderboard |
| 9 | GPT-5 mini | 51.9 | quick | Aug 2025 | Official leaderboard |
| 10 | | 51.4 | quick | Sep 2025 | Official leaderboard |
| 11 | Claude 3.5 Haiku | 50.8 | quick | Oct 2024 | Official leaderboard |
| 12 | | 49.3 | quick | Feb 2025 | Official leaderboard |
| 13 | GPT-5 nano | 45.1 | quick | Aug 2025 | Official leaderboard |
| 14 | GPT-4o mini | 39.6 | quick | Jul 2024 | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | GPT-5.5 | 65.6 | quick | Apr 2026 | Official leaderboard |
| 2 | Claude Sonnet 4.5 | 65.3 | quick | Sep 2025 | Official leaderboard |
| 3 | | 63.5 | quick | Feb 2026 | Official leaderboard |
| 4 | Claude Opus 4.5 | 63.3 | quick | Nov 2025 | Official leaderboard |
| 5 | Claude Haiku 4.5 | 59.1 | quick | Oct 2025 | Official leaderboard |
| 6 | GPT-4.1 | 58.3 | quick | Apr 2025 | Official leaderboard |
| 7 | GPT-5 mini | 51.9 | quick | Aug 2025 | Official leaderboard |
| 8 | | 51.4 | quick | Sep 2025 | Official leaderboard |
| 9 | Claude 3.5 Haiku | 50.8 | quick | Oct 2024 | Official leaderboard |
| 10 | | 49.3 | quick | Feb 2025 | Official leaderboard |
| 11 | GPT-5 nano | 45.1 | quick | Aug 2025 | Official leaderboard |
| 12 | GPT-4o mini | 39.6 | quick | Jul 2024 | Official leaderboard |
Rankings by part
We read 6 parts from the benchmark’s paper, each ranking models on one part of the test, such as M1: Task Outcome Accuracy, M2: Tool Actioning Reliability and M3: Planning & Workflow. GPT-5.5 leads 3 of them, and 2 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.
M1: Task Outcome Accuracy
12 models we track have a score here.
- 1 GPT-5.5 83.0
- 2
Gemini 3.1 Pro ↑1 78.5
- 3 Claude Opus 4.5 ↑1 74.3
- 4 Claude Sonnet 4.5 ↓2 69.9
- 5 GPT-4.1 ↑1 69.9
- 6 Claude Haiku 4.5 ↓1 69.0
- 7
Gemini 2.5 Flash Sep 2025 ↑1 59.3
- 8 GPT-5 mini ↓1 59.1
- 9
Gemini 2.0 Flash Feb 2025 ↑1 48.6
- 10 Claude 3.5 Haiku ↓1 48.2
M2: Tool Actioning Reliability
12 models we track have a score here.
- 1 Claude Sonnet 4.5 ↑1 87.2
- 2 GPT-4.1 ↑4 81.6
- 3 GPT-5.5 ↓2 79.6
- 4
Gemini 3.1 Pro ↓1 76.8
- 5 Claude 3.5 Haiku ↑4 75.8
- 6 Claude Opus 4.5 ↓2 75.6
- 7 Claude Haiku 4.5 ↓2 74.0
- 8
Gemini 2.0 Flash Feb 2025 ↑2 72.5
- 9 GPT-5 nano ↑2 72.0
- 10 GPT-5 mini ↓3 67.7
M3: Planning & Workflow
12 models we track have a score here.
- 1 GPT-5.5 30.9
- 2 Claude Opus 4.5 ↑2 27.9
- 3 Claude Sonnet 4.5 ↓1 27.5
- 4 GPT-4.1 ↑2 27.4
- 5
Gemini 2.5 Flash Sep 2025 ↑3 24.7
- 6 Claude Haiku 4.5 ↓1 24.1
- 7 GPT-5 mini 21.8
- 8
Gemini 2.0 Flash Feb 2025 ↑2 20.6
- 9
Gemini 3.1 Pro ↓6 19.6
- 10 Claude 3.5 Haiku ↓1 18.4
M4: Retrieval & Evidence Grounding
12 models we track have a score here.
- 1 GPT-5.5 54.9
- 2
Gemini 3.1 Pro ↑1 54.4
- 3 Claude Sonnet 4.5 ↓1 53.0
- 4 Claude Opus 4.5 52.1
- 5 Claude Haiku 4.5 48.6
- 6 GPT-4.1 45.0
- 7 Claude 3.5 Haiku ↑2 44.4
- 8 GPT-5 nano ↑3 43.3
- 9 GPT-5 mini ↓2 43.0
- 10
Gemini 2.0 Flash Feb 2025 42.5
M5: Safety, Abstention & Policy
12 models we track have a score here.
- 1 Claude Sonnet 4.5 ↑1 80.3
- 2 Claude Opus 4.5 ↑2 75.6
- 3
Gemini 3.1 Pro 75.6
- 4 Claude Haiku 4.5 ↑1 65.7
- 5 GPT-5.5 ↓4 63.5
- 6 Claude 3.5 Haiku ↑3 58.6
- 7
Gemini 2.5 Flash Sep 2025 ↑1 54.6
- 8 GPT-5 mini ↓1 54.2
- 9
Gemini 2.0 Flash Feb 2025 ↑1 51.7
- 10 GPT-4.1 ↓4 49.0
M6: Engineering & Maintainability
12 models we track have a score here.
- 1 Claude Haiku 4.5 ↑4 61.3
- 2 Claude Opus 4.5 ↑2 59.0
- 3 Claude Sonnet 4.5 ↓1 57.3
- 4 GPT-5.5 ↓3 56.2
- 5 GPT-5 mini ↑2 56.1
- 6
Gemini 2.5 Flash Sep 2025 ↑2 55.5
- 7
Gemini 3.1 Pro ↓4 55.3
- 8 Claude 3.5 Haiku ↑1 54.4
- 9 GPT-4.1 ↓3 54.2
- 10
Gemini 2.0 Flash Feb 2025 54.0
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: The EGI-BENCH paper
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official EGI-BENCH leaderboard
No lab model card reports it yet
We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.
More in Economic work
- JobBench In use 130 tasks from 35 occupations workers most want delegated. Agent works through messy reference files, graded by fact-anchored binary rubrics. GLM-5.3 61.4%
- GDPval In use Real work tasks from 44 occupations with reference files. Experts blindly compare model deliverables with expert work. Metric is win-or-tie rate.
Seed2.1 Pro 87.9%
- GDPval-AA v2.1 In use Agents with shell and web access complete 220 GDPval gold tasks from 44 occupations. Blind pairwise comparisons give Elo ratings. Claude Opus 5.5 1,846
- AA-Briefcase v1.1 In use Version 1.1 of Artificial Analysis's multi-hour office work evaluation, with Elo ratings fitted by Crowd-BT. Claude Opus 5.5 1,822
- Agents' Last Exam v1.1 In use Agents complete 122 public long-horizon professional workflow tasks across 55 sub-industries. Deterministic graders check outcomes against hidden references. GPT-6.2 Astra 47.5%
- CoWorkBench In use Agent completes long-horizon productivity tasks across computer science, finance, law and medicine. Qwen in-house scoring (method unpublished). Claude Fable 5 75.9%
Data audit
8 of 9 checks pass, 1 warning. Look first at the Newest score check. The newest came 7 months ago, in April 2026, for GPT-5.5. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 1 warning 2 pass 2 notes
- Pass Independent runs Scores from the benchmark’s paper (14 entries).
- Pass Models scored 12 models we track, plus 2 other entries on its leaderboard.
- Warning Newest score The newest came 7 months ago, in April 2026, for GPT-5.5.
- Note Frontier models None of the top 10 has a score. The strongest tested is GPT-5.5 (#21).
- Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
- Pass Importer We read the board on arxiv.org: 12 rows for models we track and 2 other entries at the last read.
- Pass Last read November 20th, 2026.
- Pass Read worked The last read worked.
Consistency 1 pass 1 note
- Note Lab claims No model has both a lab number and an independent run.
- Pass Known issues No published audit found problems.