Benchmark Almanac

Ko-GDPval

Solar Pro 4 87.3% leads since August 2026
In use

Agents complete 170 Korean office tasks from 58 occupations in 11 industries and produce deliverable files. Rubrics grade data, content, structure, quality and format.

Upstage built it in house in the style of OpenAI's GDPval, from its OfficeVerse task pipeline.

Made by
Upstage
Released
July 2026
Kind
Tasks from real jobs
Level
Expert
Scored
Graded by a model
Answers
Agent task
Tasks
Kept private
Last in a launch post
August 2026
Last model added
August 2026
Scores refreshed
November 12th, 2026
2
Models scored in lab cards
2
Launch posts list it, from 1 lab
1
System card reports it, from 1 lab

Solar Pro 4 leads with 87.3%

Best score per model as labs report them in their own cards. Labs pick their own settings, so these compare less cleanly than independent runs.

  1. 1 Solar Pro 4 87.3%
  2. 2 Solar Open 2 86.8%

Source: Labs’ model cards

#ModelScoreSettingDateSource
1 Solar Pro 4 87.3% Aug 2026 Lab model card
2 Solar Open 2 86.8% Jul 2026 Lab model card

The benchmark's own leaderboard has 2 models. We plot every source in the charts below.

2 models have held number one in labs’ own numbers since its release in July 2026

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

Upstage
Lab claims Passed 85% in Jul 2026, by Solar Open 2
Solar Pro 4 87.3%

Source: Labs’ model cards

Solar Open 2 on the official leaderboard passed 85% in July 2026

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 Reported by the lab in its model card Best so far
0% 20% 40% 60% 80% 100% Near the ceiling (85%) Solar Pro 4 0% 20% 40% 60% 80% 100% Near the ceiling (85%) Solar Pro 4

Sources: Official Ko-GDPval leaderboard and labs’ model cards

One lab put it in a launch post in the last six months

Launch posts from every lab that has listed it, one tick per post. Tall colored ticks include it, and short gray ones leave it out.

Upstage
DateLaunch postScoreAs the post writes it
Aug 11, 2026 Solar Pro 4 87.3 Ko-GDPval* (in-house)
Jul 22, 2026 Solar Open 2 86.8 Ko-GDPval (in-house)

One system card reports it

Besides launch posts, we read the system cards, model cards and technical reports labs publish with a release. They report far more, including safety and frontier-risk tests.

DateDocumentScoreAs the document writes itSection
Jul 22, 2026 Solar Open 2 technical report 86.8 Ko-GDPval, in-house, 170 tasks, overall score (Table 4 and Table 5). Dimensions: Data 83.5, Content 89.1, Structure 96.5, Quality 79.2, Value 82.0, Format 86.7, File Format 99.8 Capabilities

2 lab claims next to the leaderboard

Each score a lab reported for its own model, beside the same model's best score on the leaderboard. Gaps mostly come from different harnesses, effort levels and task subsets. 2 of them match, which usually means one side copied the other. Every benchmark.

Model Lab says Board says Gap
Solar Open 2 86.8% Ko-GDPval (in-house) 86.8% Leaderboard Same
Solar Pro 4 87.3% Ko-GDPval* (in-house) 87.3% Leaderboard Same

More in Economic work

All 97 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
  • 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%
  • OfficeQA In use Grounded reasoning over a large corpus of U.S. Treasury Bulletin documents: locate tables and compute precise numeric answers. Claude Opus 4.7 86.3%