Benchmark Almanac

RECON

Sixtyfour XHigh 70.6% tops its leaderboard Grok 4.20 Multi-agent 36.2% leads the models we track
Not in model cards

Research agents find and verify several facts about identified people on the open web without mixing up people with similar names.

The benchmark covers 514 human-verified facts about 140 people. Sixtyfour's page lists 23 systems and reports weighted accuracy, precision and recall.

Made by
Sixtyfour
Released
April 2026
Kind
Deep research
Level
Expert
Scored
Graded by a model
Answers
Agent task
Tasks
Public, plus a private test set
Measures
Capability
Language
English
In and out
Text → Text
Interaction
Agentic
Environment
Browser
Items from
Real work and use
Refresh
One fixed set
Run by
A host
In views
Agents
Strongest model tested
GPT-6 Astra #4 on the capability index
Last model added
September 2026
Scores refreshed
November 21st, 2026
23
Entries on its own leaderboard, 14 models we track

Sixtyfour XHigh tops the leaderboard with 70.6% Grok 4.20 Multi-agent leads with 36.2% among models released by September 2026

Best score per entry from the benchmark's own leaderboard, read on November 21st, 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. The best model we track is Grok 4.20 Multi-agent, at 36.2%. This view shows only models we track, each at its best.

The newest model with a score here came out in September 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 Sixtyfour XHigh 70.6%
  2. 2 Sixtyfour High 54.3%
  3. 3 Sixtyfour Scout 48.4%
  4. 4 Sixtyfour Medium 44.7%
  5. 5 Parallel Ultra 2x 41.1%
  6. 6 Parallel Ultra 8x 37.5%
  7. 7 Grok 4.20 Multi-agent 36.2%
  8. 8 Grok 4.6 36%
  9. 9 Grok 4.3 30.9%
  10. 10 Sixtyfour Low 28.8%
  11. 11 Parallel Ultra 27.2%
  12. 12 Kimi K3 27%
  13. 13 Claude Opus 5.5 26.3%
  14. 14 GPT-6 Astra 25.1%
  15. 15 Exa agent xhigh 23.9%
  16. 16 GPT-6.1 Sol max 23%
  17. 17 GPT-5.6 Sol xhigh 20.4%
  18. 18 Gemini 3.1 Pro 13.4%
  19. 19 Claude Sonnet 5 12.6%
  20. 20 DeepSeek-V4-Flash 12.5%
  21. 21 DeepSeek-V4-Pro 12.3%
  22. 22 GLM-5.3 3.70%
  23. 23 Claude Haiku 4.5 1.60%

Source: Official RECON leaderboard

#EntryScoreSettingDateSource
1 Sixtyfour XHigh 70.6% Official leaderboard
2 Sixtyfour High 54.3% Official leaderboard
3 Sixtyfour Scout 48.4% Official leaderboard
4 Sixtyfour Medium 44.7% Official leaderboard
5 Parallel Ultra 2x 41.1% Official leaderboard
6 Parallel Ultra 8x 37.5% Official leaderboard
7 Grok 4.20 Multi-agent 36.2% Mar 2026 Official leaderboard
8 Grok 4.6 36% Aug 2026 Official leaderboard
9 Grok 4.3 30.9% Apr 2026 Official leaderboard
10 Sixtyfour Low 28.8% Official leaderboard
11 Parallel Ultra 27.2% Official leaderboard
12 Kimi K3 27% Jul 2026 Official leaderboard
13 Claude Opus 5.5 26.3% Sep 2026 Official leaderboard
14 GPT-6 Astra 25.1% Sep 2026 Official leaderboard
15 Exa agent xhigh 23.9% Official leaderboard
16 GPT-6.1 Sol 23% max Sep 2026 Official leaderboard
17 GPT-5.6 Sol 20.4% xhigh Jul 2026 Official leaderboard
18 Gemini 3.1 Pro 13.4% Feb 2026 Official leaderboard
19 Claude Sonnet 5 12.6% Jun 2026 Official leaderboard
20 DeepSeek-V4-Flash 12.5% Apr 2026 Official leaderboard
21 DeepSeek-V4-Pro 12.3% Apr 2026 Official leaderboard
22 GLM-5.3 3.70% Aug 2026 Official leaderboard
23 Claude Haiku 4.5 1.60% Oct 2025 Official leaderboard
#ModelScoreSettingDateSource
1 Grok 4.20 Multi-agent 36.2% Mar 2026 Official leaderboard
2 Grok 4.6 36% Aug 2026 Official leaderboard
3 Grok 4.3 30.9% Apr 2026 Official leaderboard
4 Kimi K3 27% Jul 2026 Official leaderboard
5 Claude Opus 5.5 26.3% Sep 2026 Official leaderboard
6 GPT-6 Astra 25.1% Sep 2026 Official leaderboard
7 GPT-6.1 Sol 23% max Sep 2026 Official leaderboard
8 GPT-5.6 Sol 20.4% xhigh Jul 2026 Official leaderboard
9 Gemini 3.1 Pro 13.4% Feb 2026 Official leaderboard
10 Claude Sonnet 5 12.6% Jun 2026 Official leaderboard
11 DeepSeek-V4-Flash 12.5% Apr 2026 Official leaderboard
12 DeepSeek-V4-Pro 12.3% Apr 2026 Official leaderboard
13 GLM-5.3 3.70% Aug 2026 Official leaderboard
14 Claude Haiku 4.5 1.60% Oct 2025 Official leaderboard

Rankings by part

We read 2 parts from the benchmark’s own leaderboard: Precision and Recall. Each ranks models on one part of the test. GPT-6 Astra leads Precision and Grok 4.20 Multi-agent leads Recall. Arrows show how far a model moved from its overall rank among the models we track.

Precision

14 models we track have a score here.

  1. 1 GPT-6 Astra ↑5 90.6%
  2. 2 GPT-6.1 Sol ↑5 85.1%
  3. 3 Claude Opus 5.5 ↑2 83.3%
  4. 4 Kimi K3 80.1%
  5. 5 Grok 4.6 ↓3 77.4%
  6. 6 Grok 4.20 Multi-agent ↓5 77%
  7. 7 Grok 4.3 ↓4 76.2%
  8. 8 DeepSeek-V4-Flash ↑3 75%
  9. 9 GPT-5.6 Sol ↓1 74.2%
  10. 10 DeepSeek-V4-Pro ↑2 74%

Number one over time

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

xAI
Own leaderboard Not tested since Sep 2026

Source: Official RECON 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 Best so far
0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) Grok 4.20 Multi-agent Gemini 3.1 Pro Claude Haiku 4.5 0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) Grok 4.20 Multi-agent Gemini 3.1 Pro

Source: Official RECON 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 Tool use

All 265 in Tool use
  • AutomationBench In use Agent completes end-to-end business workflows inside a simulated company exposing dozens of REST APIs. Graded on resulting state. Gemini 4 Neon 56.3%
  • MCP Atlas In use Multi-step workflows over real MCP servers. Scores tool-use competency on realistic tasks. Gemini 4 Argon 90.7%
  • BFCL V4 In use Models call functions and tools, including multi-turn agentic tasks with web search and memory. Checked against expected calls and outcomes. Qwen3.7-Max 75%
  • τ³-bench Banking In use Agents act as banking customer-service reps over many turns with a simulated user, following policy. Pass^1 on the final database state. Qwen 3.8 Max 55.1%
  • Toolathlon-Verified In use 108 long-horizon tasks across 32 apps through about 600 MCP tools. Execution-based checkers inspect final environment state. GLM-5.3-Flash 78.4%
  • DeepSearchQA In use Questions that need multi-step search and synthesis. Final answers graded against references. Kimi K3 95%

Data audit

All 11 checks pass. Every benchmark

Catalog record 2 pass
  • Pass Task access Public tasks, plus a private test set.
  • Pass Confidence Checked against a primary source.
Scores 5 pass
  • Pass Independent runs Scores from the official leaderboard (23 entries).
  • Pass Models scored 14 models we track, plus 9 other entries on its leaderboard.
  • Pass Newest score The newest is from September 2026, for GPT-6.1 Sol.
  • Pass Frontier models 3 of the top 10 have a score, led by GPT-6 Astra (#4).
  • Pass Headroom The best is 36.2%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on sixtyfour.ai: 14 rows for models we track and 9 other entries at the last read.
  • Pass Last read November 21st, 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