Benchmark Almanac

FACTS Search V2

GPT-6 Astra 89.9% leads since September 2026
Not in model cards

Models answer information-seeking questions with a search tool and must combine what they find correctly. Second version of FACTS Search on Kaggle.

Kaggle published version 2 of the board on July 27th, 2026 and renamed it FACTS Search V2. It listed 23 models on November 1st, 2026. No changelog was found.

Released
July 2026
Kind
Short factual questions
Level
University
Scored
Graded by a model
Answers
Short answer
Tasks
Public, plus a private test set
Strongest model tested
GPT-6 Astra #3 on the capability index
Last model added
October 2026
Scores refreshed
November 12th, 2026
22
Entries on its own leaderboard, 21 models we track

GPT-6 Astra leads with 89.9%

Best score per entry from the benchmark's own leaderboard, read on November 12th, 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.

22 entries. Scroll sideways for the rest.

  1. 89.9 GPT-6 Astra
  2. 88.5 GPT-5.6 Sol
  3. 88.2 GPT-6.1 Sol
  4. 86.5 Claude Opus 5.5
  5. 86.2 GPT-6 Sol
  6. 85.3 Claude Opus 5
  7. 84.0 Gemini 4 Argon
  8. 83.7 Gemini 3.1 Pro
  9. 81.7 GPT-5.6 Terra
  10. 81.4 GPT-6 Luna
  11. 80.9 Gemini 3 Flash
  12. 80.9 Gemini 3.7 Flash
  13. 79.6 Claude Opus 4.8
  14. 79.4 Grok 4.5
  15. 79.3 GPT-5.6 Luna
  16. 77.5 Claude Haiku 5.5
  17. 76.3 Gemini 3.8 Flash
  18. 68.2 Gemini 3.6 Flash
  19. 64.3 Gemini 3.5 Flash
  20. 63.4 Gemini 3.5 Flash-Lite
  21. 60.3 Gemma 4 26B A4B
  22. 53.4 Qwen 3 235B A22B Instruct 2506
  1. 1 GPT-6 Astra 89.9%
  2. 2 GPT-5.6 Sol 88.5%
  3. 3 GPT-6.1 Sol 88.2%
  4. 4 Claude Opus 5.5 86.5%
  5. 5 GPT-6 Sol 86.2%
  6. 6 Claude Opus 5 85.3%
  7. 7 Gemini 4 Argon 84%
  8. 8 Gemini 3.1 Pro 83.7%
  9. 9 GPT-5.6 Terra 81.7%
  10. 10 GPT-6 Luna 81.4%
  11. 11 Gemini 3 Flash 80.9%
  12. 12 Gemini 3.7 Flash 80.9%
  13. 13 Claude Opus 4.8 79.6%
  14. 14 Grok 4.5 79.4%
  15. 15 GPT-5.6 Luna 79.3%
  16. 16 Claude Haiku 5.5 77.5%
  17. 17 Gemini 3.8 Flash 76.3%
  18. 18 Gemini 3.6 Flash 68.2%
  19. 19 Gemini 3.5 Flash 64.3%
  20. 20 Gemini 3.5 Flash-Lite 63.4%
  21. 21 Gemma 4 26B A4B 60.3%
  22. 22 Qwen 3 235B A22B Instruct 2506 53.4%

Source: Official FACTS Search V2 leaderboard

#EntryScoreSettingDateSource
1 GPT-6 Astra 89.9% Sep 2026 Official leaderboard
2 GPT-5.6 Sol 88.5% Jul 2026 Official leaderboard
3 GPT-6.1 Sol 88.2% Sep 2026 Official leaderboard
4 Claude Opus 5.5 86.5% Sep 2026 Official leaderboard
5 GPT-6 Sol 86.2% Sep 2026 Official leaderboard
6 Claude Opus 5 85.3% Jul 2026 Official leaderboard
7 Gemini 4 Argon 84% Sep 2026 Official leaderboard
8 Gemini 3.1 Pro 83.7% Feb 2026 Official leaderboard
9 GPT-5.6 Terra 81.7% Jul 2026 Official leaderboard
10 GPT-6 Luna 81.4% Sep 2026 Official leaderboard
11 Gemini 3 Flash 80.9% Dec 2025 Official leaderboard
12 Gemini 3.7 Flash 80.9% Aug 2026 Official leaderboard
13 Claude Opus 4.8 79.6% May 2026 Official leaderboard
14 Grok 4.5 79.4% Jul 2026 Official leaderboard
15 GPT-5.6 Luna 79.3% Jul 2026 Official leaderboard
16 Claude Haiku 5.5 77.5% Oct 2026 Official leaderboard
17 Gemini 3.8 Flash 76.3% Sep 2026 Official leaderboard
18 Gemini 3.6 Flash 68.2% Jul 2026 Official leaderboard
19 Gemini 3.5 Flash 64.3% May 2026 Official leaderboard
20 Gemini 3.5 Flash-Lite 63.4% Jul 2026 Official leaderboard
21 Gemma 4 26B A4B 60.3% Apr 2026 Official leaderboard
22 Qwen 3 235B A22B Instruct 2506 53.4% Jul 2026 Official leaderboard

3 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.

Google DeepMind OpenAI
Own leaderboard Passed 85% in Jul 2026, by GPT-5.6 Sol
GPT-6 Astra 89.9%

Source: Official FACTS Search V2 leaderboard

GPT-5.6 Sol 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 Best so far
0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) GPT-6 Astra Gemini 3.1 Pro 0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) GPT-6 Astra Gemini 3 Flash

Source: Official FACTS Search V2 leaderboard

No lab model card reports it yet

We read the headline results of 522 launch posts from 52 labs, and none of them list this benchmark.

Versions

We switch the top row to each new version when it comes out. Open it for one row per version. An old version stays in play while people still test it, so whoever first takes it past 85% gets credit for closing it out.

OpenAI Google DeepMind
FACTS Search V2 Passed 85% in Jul 2026, by GPT-5.6 Sol
GPT-6 Astra 89.9%
FACTS Search Passed 85% in Feb 2026, by Gemini 3.1 Pro
Gemini 3.1 Pro 85.6%
FACTS Search V2 Passed 85% in Jul 2026, by GPT-5.6 Sol
GPT-6 Astra 89.9%

Source: Official leaderboards

Replaces FACTS Search

More in Factuality

All 74 in Factuality
  • SimpleVQA In use Models answer 2,025 short factual questions about images across 9 tasks and 9 topics. An LLM judge grades answers against fixed references. Qwen3.7-Plus 81.7%
  • HallusionBench In use Models answer 1,129 yes or no questions about 346 images, built in pairs to expose language hallucination and visual illusion. Scored by accuracy. Keye-VL-2.0-30B-A3B 73.5%
  • SimpleQA Verified In use 1,000 human-verified, de-duplicated short-answer factual questions measuring parametric knowledge. Gemini 3.1 Pro 77.5%
  • Kimi Hallu Bench In use Models answer hard prompts drawn from practical work tasks. Fact-checking agents score the share of claims that are hallucinated. GPT-6 Astra 4.20%
  • Internal adversarially-designed factuality evaluations In use Models answer adversarial questions in nine categories such as history, math and code. Scored on agreement with human ideal responses. GPT-6 Luna 7.60%
  • AA-Omniscience v1.1 In use Models answer 3,000 knowledge questions in six domains without tools and may abstain. The index adds correct answers and subtracts wrong ones. Inkling 2 Small 16.1