Benchmark Almanac

Prosus Vending Bench v2

Claude Opus 5 4497.2 leads since September 2026
Not in model cards

Agents run six simulated vending machines across three locations for 30 days. They order stock, set prices, and collect cash. The score measures final bank balance in euros.

Prosus keeps a ranked v2 table for five models on one seed. The public simulator uses v3 rules. Prosus removed the original trial files.

Made by
Prosus AI
Released
September 2026
Kind
Running a business
Level
University
Scored
Checked automatically
Answers
Agent task
Measures
Capability
Language
English
In and out
Text → Actions, Text
Interaction
Agentic
Environment
A simulator
Items from
Generated
Run by
A host
Strongest model tested
Claude Opus 5 #11 on the capability index
Last model added
July 2026
Scores refreshed
November 20th, 2026
5
Models scored on its own leaderboard

Claude Opus 5 leads with 4497.2 among models released by July 2026

Best score per model from the benchmark's own leaderboard, read on November 20th, 2026. Where a model ran with several settings or harnesses, we keep its best.

The newest model with a score here came out in July 2026, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

  1. 1 Claude Opus 5 30 days, seed 1 4497.2
  2. 2 GPT-5.6 Sol 30 days, seed 1 3357.8
  3. 3 Claude Sonnet 5 30 days, seed 1 3289.1
  4. 4 GPT-5.6 Luna 30 days, seed 1 1601.6
  5. 5 Claude Haiku 4.5 30 days, seed 1 1337.7

Source: Official Prosus Vending Bench v2 leaderboard

#ModelScoreSettingDateSource
1 Claude Opus 5 4497.2 30 days, seed 1, In-process MCP client Jul 2026 Official leaderboard
2 GPT-5.6 Sol 3357.8 30 days, seed 1, In-process MCP client Jul 2026 Official leaderboard
3 Claude Sonnet 5 3289.1 30 days, seed 1, In-process MCP client Jun 2026 Official leaderboard
4 GPT-5.6 Luna 1601.6 30 days, seed 1, In-process MCP client Jul 2026 Official leaderboard
5 Claude Haiku 4.5 1337.7 30 days, seed 1, In-process MCP client Oct 2025 Official leaderboard

Number one over time

Lab claims count from the day the card came out, independent runs from the model’s release.

Anthropic
Own leaderboard Not tested since Sep 2026
Claude Opus 5 4497.2

Source: Official Prosus Vending Bench v2 leaderboard

Scores over time

All the scores we have, one dot each, so a model tested at several settings appears more than once.

Plot with other benchmarks
On the benchmark’s own leaderboard
1000.0 2000.0 3000.0 4000.0 5000.0 2026 1000.0 2000.0 3000.0 4000.0 5000.0 2026

Source: Official Prosus Vending Bench v2 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

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

8 of 9 checks pass, 1 warning. Look first at the Task access check. We don’t know whether the tasks are public. Every benchmark

Catalog record 1 warning 1 pass
  • Warning Task access We don’t know whether the tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 3 pass 2 notes
  • Pass Independent runs Scores from the official leaderboard (5 entries).
  • Pass Models scored 5 models we track.
  • Pass Newest score The newest is from July 2026, for Claude Opus 5.
  • Note Frontier models None of the top 10 has a score. The strongest tested is Claude Opus 5 (#11).
  • Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
  • Pass Importer We read the board on github.com: 5 rows for models we track 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