Benchmark Almanac

Groundtruth Dynamic Benchmarking (Geology Edition)

Kimi K3 8.81 leads since July 2026
Not in model cards

Models answer questions from geological reports. A judge scores each answer from zero to ten against a rubric tied to passages in the source documents.

Eigenform publishes eight models across three sets of 50 questions. Submitters report their own scores.

Made by
Eigenform
Released
July 2026
Kind
Reading comprehension
Level
Expert
Scored
Graded by a model
Answers
Open-ended
Tasks
All public
Measures
Capability
Language
English
In and out
Text → Text
Interaction
Single turn
Items from
Real work and use
Refresh
One fixed set
Run by
Anyone
Strongest model tested
GPT-5.6 Sol #14 on the capability index
Last model added
August 2026
Scores refreshed
November 20th, 2026
8
Entries on its own leaderboard, 7 models we track

Kimi K3 leads with 8.81 among models released by August 2026 Kimi K3 leads with 8.81 among models released by August 2026

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.

The newest model with a score here came out in August 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 Kimi K3 8.81
  2. 2 Claude Sonnet 5 8.62
  3. 3 GPT-5.6 Sol 8.50
  4. 4 OxAlpha 8.44
  5. 5 DeepSeek-V4-Pro 7.79
  6. 6 Grok 4.6 7.73
  7. 7 Gemini 3.1 Pro 7.55
  8. 8 GLM-4.7 6.81

Source: Official Groundtruth Dynamic Benchmarking (Geology Edition) leaderboard

#EntryScore ±SettingDateSource
1 Kimi K3 8.81 8.7 to 9 Jul 2026 Official leaderboard
2 Claude Sonnet 5 8.62 8.4 to 8.8 Jun 2026 Official leaderboard
3 GPT-5.6 Sol 8.50 ±0.2 Jul 2026 Official leaderboard
4 OxAlpha 8.44 8.1 to 8.7 Official leaderboard
5 DeepSeek-V4-Pro 7.79 ±0.2 Apr 2026 Official leaderboard
6 Grok 4.6 7.73 ±0.2 Aug 2026 Official leaderboard
7 Gemini 3.1 Pro 7.55 7.3 to 7.8 Feb 2026 Official leaderboard
8 GLM-4.7 6.81 6.5 to 7.1 Dec 2025 Official leaderboard
#ModelScore ±SettingDateSource
1 Kimi K3 8.81 8.7 to 9 Jul 2026 Official leaderboard
2 Claude Sonnet 5 8.62 8.4 to 8.8 Jun 2026 Official leaderboard
3 GPT-5.6 Sol 8.50 ±0.2 Jul 2026 Official leaderboard
4 DeepSeek-V4-Pro 7.79 ±0.2 Apr 2026 Official leaderboard
5 Grok 4.6 7.73 ±0.2 Aug 2026 Official leaderboard
6 Gemini 3.1 Pro 7.55 7.3 to 7.8 Feb 2026 Official leaderboard
7 GLM-4.7 6.81 6.5 to 7.1 Dec 2025 Official leaderboard

Rankings by part

We read 15 parts from the benchmark’s own leaderboard in 3 groups: technical, wamex and usgs. Each ranks models on one part of the test. Kimi K3 leads 10 of them, and Claude Sonnet 5 leads the other 5. Arrows show how far a model moved from its overall rank among the models we track.

Technical

7 models we track have a score here.

  1. 1 Kimi K3 8.62
  2. 2 Claude Sonnet 5 8.46
  3. 3 GPT-5.6 Sol 8.36
  4. 4 Gemini 3.1 Pro ↑2 7.62
  5. 5 Grok 4.6 7.62
  6. 6 DeepSeek-V4-Pro ↓2 7.34
  7. 7 GLM-4.7 6.76

2 models have held number one on the official leaderboard since its release in July 2026

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

Anthropic Moonshot AI
Own leaderboard Not tested since Aug 2026
Kimi K3 8.81

Source: Official Groundtruth Dynamic Benchmarking (Geology Edition) 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
6.50 7.00 7.50 8.00 8.50 9.00 2026 6.50 7.00 7.50 8.00 8.50 9.00 2026

Source: Official Groundtruth Dynamic Benchmarking (Geology Edition) 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 Knowledge

All 332 in Knowledge
  • MMLU-Pro In use Harder 10-option version of MMLU with reasoning-focused questions across 14 subjects. Intern-S2-397B 89.8%
  • MMLU In use 57-subject multiple-choice exam questions from elementary to professional level. Scored by accuracy. Seed1.8 92.3%
  • SuperGPQA In use 26,529 graduate-level multiple-choice questions across 285 disciplines. Accuracy. Qwen 3.6 Max (Preview) 73.9%
  • WideSearch In use 200 bilingual broad info-seeking tasks: agent collects many atomic facts into a table. Scored by success rate and row/item F1. Tencent Hy4 preview 83.9%
  • BBEH micro average In use Models solve harder replacements for each of the 23 BIG-Bench Hard tasks, 4,520 examples in total. The score is the share of all examples answered correctly. Gemma 4 31B IT 74.4%
  • HLE-Diamond In use Models answer 1,000 cleaned expert questions from Humanity's Last Exam, 500 on reasoning and 500 on knowledge, closed book. Graded on exact answers. Claude Fable 5.2 77.9%

Data audit

All 9 checks pass. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 3 pass 2 notes
  • Pass Independent runs Scores from the official leaderboard (8 entries).
  • Pass Models scored 7 models we track, plus 1 other entry on its leaderboard.
  • Pass Newest score The newest is from August 2026, for Grok 4.6.
  • Note Frontier models None of the top 10 has a score. The strongest tested is GPT-5.6 Sol (#14).
  • Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
  • Pass Importer We read the board on eigenform.ai: 7 rows for models we track and 1 other entry 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