Benchmark Almanac

MANDOS Run 003

Not in model cards
Version Run 003

Models make decisions during seven industrial incidents and four fictional scenarios. A model judge checks their decisions, including whether they seek expert advice.

RelayForge publishes outcomes for five distinct operators on the seven incident scenarios. Two configurations use Grok as the operator with different judges.

Made by
RelayForge
Released
June 2026
Kind
Tasks from real jobs
Level
Expert
Scored
Graded by a model
Answers
Agent task
Tasks
Kept private
Measures
Propensity
Behavior
Hallucination
Language
English
In and out
Text → Actions, Text
Interaction
Agentic
Environment
A simulator
Run by
A host
Occupations
Production
Last model added
March 2025
Scores refreshed
November 20th, 2026
5
Entries on its own leaderboard, 1 model we track

GPT-4o leads with 100% GPT-4o leads with 100%

Best score per entry from the benchmark's own leaderboard, read on November 20th, 2026. 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.

  1. 1 Gemini 2.5 100%
  2. 2 Grok Claude Opus judge 100%
  3. 3 Llama 70B 100%
  4. 4 GPT-4o Mar 2025 100%
  5. 5 Qwen 32B 85.7%

Source: Official MANDOS Run 003 leaderboard

Source: Official MANDOS Run 003 leaderboard

#EntryScoreSettingDateSource
1 Gemini 2.5 100% Official leaderboard
2 Grok 100% Claude Opus judge Official leaderboard
3 Llama 70B 100% Official leaderboard
4 GPT-4o Mar 2025 100% Mar 2025 Official leaderboard
5 Qwen 32B 85.7% Official leaderboard
#ModelScoreSettingDateSource
1 GPT-4o Mar 2025 100% Mar 2025 Official leaderboard

GPT-4o on the official leaderboard passed 85% in March 2025

Models released before it came out already scored that high. All the scores we have, one dot each, so a model tested at several settings appears more than once.

On the benchmark’s own leaderboard
0% 20% 40% 60% 80% 100% 2025 Near the ceiling (85%) GPT-4o 0% 20% 40% 60% 80% 100% 2025 Near the ceiling (85%) GPT-4o

Source: Official MANDOS Run 003 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.

Versions

More in Economic work

All 210 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

All 8 checks pass. Every benchmark

Catalog record 2 pass
  • Pass Task access The tasks are private.
  • Pass Confidence Checked against a primary source.
Scores 2 pass 3 notes
  • Pass Independent runs Scores from the official leaderboard (5 entries).
  • Pass Models scored 1 model we track, plus 4 other entries on its leaderboard.
  • Note Newest score The newest came 20 months ago, in March 2025, for GPT-4o. That’s expected for a test replaced by MANDOS Session 2.
  • Note Frontier models None of its models are on the capabilities index.
  • Note Headroom Saturated. The best is 100%, past 85% since March 2025.
Leaderboard import 3 pass
  • Pass Importer We read the board on relayforge.tools: 1 row for models we track and 4 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.

How we check