Benchmark Almanac

Implicit Intelligence

Claude Opus 4.6 53.2% leads since February 2026
Not in model cards

Agents respond to requests whose unstated requirements affect safety, privacy, or accessibility. Labelbox reports the share of scenarios each model passes.

Labelbox publishes the results on its leaderboard.

Made by
Labelbox
Kind
General assistants
Level
Everyday
Scored
Settled by events
Answers
Agent task
Tasks
Kept private
Measures
Capability
Risk
Privacy
Language
English
In and out
Text → Actions
Interaction
Agentic
Environment
Phone apps, A simulator
Items from
Generated
Refresh
One fixed set
Run by
A host
Strongest model tested
GPT-5.4 Pro #19 on the capability index
Last model added
April 2026
Scores refreshed
November 20th, 2026
20
Models scored on its own leaderboard

Claude Opus 4.6 leads with 53.2% among models released by April 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. Every run is in the next chart.

The newest model with a score here came out in April 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 4.6 high 53.2%
  2. 2 GPT-5.2 Pro 48.3%
  3. 3 Claude Opus 4.7 high 45.9%
  4. 4 GPT-5 44.9%
  5. 5 GPT-5.4 Pro 43.9%
  6. 6 Claude Opus 4.5 high 41%
  7. 7 Gemini 3 Pro 38.5%
  8. 8 GPT-5.2 high 35.1%
  9. 9 DeepSeek-R1 May 2025 34.6%
  10. 10 Gemini 3 Flash 30.2%
  11. 11 Claude Sonnet 4.5 28.3%
  12. 12 DeepSeek-V3.1 27.3%
  13. 13 GPT-5.4 27.3%
  14. 14 GPT-5.1 20.5%
  15. 15 GPT-4.1 18.5%
  16. 16 Llama 4 Maverick 18%
  17. 17 gpt-oss-120b 16.1%
  18. 18 Llama 4 Scout 11.2%
  19. 19 gpt-oss-20b 9.80%
  20. 20 Gemma 3n E4B 4.90%

Source: Official Implicit Intelligence leaderboard

#ModelScore ±SettingDateSource
1 Claude Opus 4.6 53.2% ±6.6 high Feb 2026 Official leaderboard
2 GPT-5.2 Pro 48.3% ±6.8 Dec 2025 Official leaderboard
3 Claude Opus 4.7 45.9% ±6.8 high Apr 2026 Official leaderboard
4 GPT-5 44.9% ±6.8 Aug 2025 Official leaderboard
5 GPT-5.4 Pro 43.9% ±6.8 Mar 2026 Official leaderboard
6 Claude Opus 4.5 41% ±6.8 high Nov 2025 Official leaderboard
7 Gemini 3 Pro 38.5% ±6.3 Nov 2025 Official leaderboard
8 GPT-5.2 35.1% ±6.6 high Dec 2025 Official leaderboard
9 DeepSeek-R1 May 2025 34.6% ±6.6 May 2025 Official leaderboard
10 Gemini 3 Flash 30.2% ±6.1 Dec 2025 Official leaderboard
11 Claude Sonnet 4.5 28.3% ±6.1 Sep 2025 Official leaderboard
12 DeepSeek-V3.1 27.3% ±6.1 Aug 2025 Official leaderboard
13 GPT-5.4 27.3% ±6.1 Mar 2026 Official leaderboard
14 GPT-5.1 20.5% ±5.6 Nov 2025 Official leaderboard
15 GPT-4.1 18.5% ±5.4 Apr 2025 Official leaderboard
16 Llama 4 Maverick 18% ±5.1 Apr 2025 Official leaderboard
17 gpt-oss-120b 16.1% ±5.1 Aug 2025 Official leaderboard
18 Llama 4 Scout 11.2% ±4.1 Apr 2025 Official leaderboard
19 gpt-oss-20b 9.80% ±4.1 Aug 2025 Official leaderboard
20 Gemma 3n E4B 4.90% ±3.2 Jun 2025 Official leaderboard

Rankings by part

We read 4 parts from the benchmark’s own leaderboard, each ranking models on one part of the test, such as Implicit Reasoning, Catastrophic Risk and Privacy & Security. Claude Opus 4.6 leads 3 of them, and GPT-5.2 Pro leads the other one. Arrows show how far a model moved from its overall rank among the models we track.

Implicit Reasoning

20 models we track have a score here.

  1. 1 GPT-5.2 Pro ↑1 51.4%
  2. 2 Claude Opus 4.6 ↓1 50%
  3. 3 Gemini 3 Pro ↑4 45.7%
  4. 4 GPT-5.4 Pro ↑1 44.3%
  5. 5 GPT-5 ↓1 42.9%
  6. 6 Claude Opus 4.7 ↓3 38.6%
  7. 7 Claude Opus 4.5 ↓1 35.7%
  8. 8 Gemini 3 Flash ↑2 32.9%
  9. 9 DeepSeek-R1 May 2025 31.4%
  10. 10 DeepSeek-V3.1 ↑2 31.4%

Scores at each effort level

Models the leaderboard ran at two or more effort levels. We don’t have enough cost data here to say what the extra effort costs.

Lowest effort to highest, light to dark
Claude Opus 4.6
high
53.2%
GPT-5.2 Pro
high
48.3%
Claude Opus 4.7
high
45.9%
GPT-5
high
44.9%
GPT-5.4 Pro
xhigh
43.9%
Claude Opus 4.5
high
41%
Gemini 3 Pro
high
38.5%
GPT-5.2
high
35.1%
Claude Sonnet 4.5
high
28.3%
GPT-5.4
xhigh
27.3%

Source: Official Implicit Intelligence leaderboard

6 models have held number one on the official leaderboard since 2023

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

Meta OpenAI DeepSeek Anthropic
Own leaderboard Not tested since Apr 2026

Source: Official Implicit Intelligence 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% 2025 2026 Near the ceiling (85%) Claude Opus 4.6 GPT-5.2 Pro GPT-5 DeepSeek-R1 GPT-4.1 0% 20% 40% 60% 80% 100% 2025 2026 Near the ceiling (85%) Claude Opus 4.6 GPT-5

Source: Official Implicit Intelligence 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

9 of 10 checks pass, 1 warning. Look first at the Newest score check. The newest came 7 months ago, in April 2026, for Claude Opus 4.7. Every benchmark

Catalog record 2 pass
  • Pass Task access The tasks are private.
  • Pass Confidence Checked against a primary source.
Scores 1 warning 3 pass 1 note
  • Pass Independent runs Scores from the official leaderboard (20 entries).
  • Pass Models scored 20 models we track.
  • Warning Newest score The newest came 7 months ago, in April 2026, for Claude Opus 4.7.
  • Note Frontier models None of the top 10 has a score. The strongest tested is GPT-5.4 Pro (#19).
  • Pass Headroom The best is 53.2%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on labelbox.com: 30 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