Implicit Intelligence
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
- Interaction
- Agentic
- Environment
- Phone apps, A simulator
- Items from
- Generated
- Refresh
- One fixed set
- Run by
- A host
- In views
- Agents, Robotics, Content safety
- Strongest model tested
- GPT-5.4 Pro #19 on the capability index
- Last model added
- April 2026
- Scores refreshed
- November 20th, 2026
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:
- GPT-6.2 Astra
- Claude Fable 5.2
-
Gemini 4 Carbon
- GPT-6 Astra
- and 15 more
- 53.2 Claude Opus 4.6
- 48.3 GPT-5.2 Pro
- 45.9 Claude Opus 4.7
- 44.9 GPT-5
- 43.9 GPT-5.4 Pro
- 41.0 Claude Opus 4.5
- 38.5
Gemini 3 Pro
- 35.1 GPT-5.2
- 34.6
DeepSeek-R1
- 30.2
Gemini 3 Flash
- 28.3 Claude Sonnet 4.5
- 27.3
DeepSeek-V3.1
- 27.3 GPT-5.4
- 20.5 GPT-5.1
- 18.5 GPT-4.1
- 18.0
Llama 4 Maverick
- 16.1 gpt-oss-120b
- 11.2
Llama 4 Scout
- 9.80 gpt-oss-20b
- 4.90
Gemma 3n E4B
- 1 Claude Opus 4.6 high 53.2%
- 2 GPT-5.2 Pro 48.3%
- 3 Claude Opus 4.7 high 45.9%
- 4 GPT-5 44.9%
- 5 GPT-5.4 Pro 43.9%
- 6 Claude Opus 4.5 high 41%
- 7
Gemini 3 Pro 38.5%
- 8 GPT-5.2 high 35.1%
- 9
DeepSeek-R1 May 2025 34.6%
- 10
Gemini 3 Flash 30.2%
- 11 Claude Sonnet 4.5 28.3%
- 12
DeepSeek-V3.1 27.3%
- 13 GPT-5.4 27.3%
- 14 GPT-5.1 20.5%
- 15 GPT-4.1 18.5%
- 16
Llama 4 Maverick 18%
- 17 gpt-oss-120b 16.1%
- 18
Llama 4 Scout 11.2%
- 19 gpt-oss-20b 9.80%
- 20
Gemma 3n E4B 4.90%
Source: Official Implicit Intelligence leaderboard
| # | Model | Score ± | Setting | Date | Source |
|---|---|---|---|---|---|
| 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 | | 38.5% ±6.3 | Nov 2025 | Official leaderboard | |
| 8 | GPT-5.2 | 35.1% ±6.6 | high | Dec 2025 | Official leaderboard |
| 9 | | 34.6% ±6.6 | May 2025 | Official leaderboard | |
| 10 | | 30.2% ±6.1 | Dec 2025 | Official leaderboard | |
| 11 | Claude Sonnet 4.5 | 28.3% ±6.1 | Sep 2025 | Official leaderboard | |
| 12 | | 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 | | 18% ±5.1 | Apr 2025 | Official leaderboard | |
| 17 | gpt-oss-120b | 16.1% ±5.1 | Aug 2025 | Official leaderboard | |
| 18 | | 11.2% ±4.1 | Apr 2025 | Official leaderboard | |
| 19 | gpt-oss-20b | 9.80% ±4.1 | Aug 2025 | Official leaderboard | |
| 20 | | 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 GPT-5.2 Pro ↑1 51.4%
- 2 Claude Opus 4.6 ↓1 50%
- 3
Gemini 3 Pro ↑4 45.7%
- 4 GPT-5.4 Pro ↑1 44.3%
- 5 GPT-5 ↓1 42.9%
- 6 Claude Opus 4.7 ↓3 38.6%
- 7 Claude Opus 4.5 ↓1 35.7%
- 8
Gemini 3 Flash ↑2 32.9%
- 9
DeepSeek-R1 May 2025 31.4%
- 10
DeepSeek-V3.1 ↑2 31.4%
Catastrophic Risk
20 models we track have a score here.
- 1 Claude Opus 4.6 57.1%
- 2 Claude Opus 4.7 ↑1 55.4%
- 3 Claude Opus 4.5 ↑3 50%
- 4 GPT-5 48.2%
- 5 GPT-5.2 Pro ↓3 48.2%
- 6 GPT-5.4 Pro ↓1 46.4%
- 7
Gemini 3 Pro 41.1%
- 8 GPT-5.2 39.3%
- 9
Gemini 3 Flash ↑1 37.5%
- 10 Claude Sonnet 4.5 ↑1 35.7%
Privacy & Security
20 models we track have a score here.
- 1 Claude Opus 4.6 58.7%
- 2 GPT-5.2 Pro 52.2%
- 3 Claude Opus 4.7 52.2%
- 4 GPT-5.4 Pro ↑1 50%
- 5
DeepSeek-R1 May 2025 ↑4 45.7%
- 6 GPT-5 ↓2 43.5%
- 7 Claude Opus 4.5 ↓1 43.5%
- 8 GPT-5.2 43.5%
- 9 GPT-5.4 ↑4 32.6%
- 10
Gemini 3 Pro ↓3 30.4%
Accessibility
20 models we track have a score here.
- 1 Claude Opus 4.6 51.5%
- 2 GPT-5 ↑2 48.5%
- 3 Claude Opus 4.5 ↑3 42.4%
- 4 GPT-5.2 Pro ↓2 42.4%
- 5
Gemini 3 Pro ↑2 39.4%
- 6 GPT-5.2 ↑2 39.4%
- 7 GPT-5.4 Pro ↓2 39.4%
- 8 Claude Opus 4.7 ↓5 39.4%
- 9 Claude Sonnet 4.5 ↑2 36.4%
- 10 GPT-5.4 ↑3 30.3%
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.
Source: Official Implicit Intelligence leaderboard
| Model | Effort | Harness | Score | Cost per task | Output tokens |
|---|---|---|---|---|---|
| Claude Opus 4.6 | high | 53.2% | |||
| Claude Opus 4.6 | 52.7% | ||||
| GPT-5.2 Pro | 48.3% | ||||
| GPT-5.2 Pro | high | 47.3% | |||
| Claude Opus 4.7 | high | 45.9% | |||
| GPT-5 | 44.9% | ||||
| GPT-5.4 Pro | 43.9% | ||||
| GPT-5.4 Pro | xhigh | 43.4% | |||
| Claude Opus 4.7 | 42.4% | ||||
| GPT-5 | high | 41.5% | |||
| Claude Opus 4.5 | high | 41% | |||
| Claude Opus 4.5 | 39.5% | ||||
| 38.5% | |||||
| high | 37.6% | ||||
| GPT-5.2 | high | 35.1% | |||
| 34.6% | |||||
| GPT-5.2 | 33.7% | ||||
| 30.2% | |||||
| Claude Sonnet 4.5 | 28.3% | ||||
| Claude Sonnet 4.5 | high | 27.8% | |||
| 27.3% | |||||
| GPT-5.4 | 27.3% | ||||
| GPT-5.4 | xhigh | 25.9% | |||
| GPT-5.1 | 20.5% | ||||
| GPT-4.1 | 18.5% | ||||
| 18% | |||||
| gpt-oss-120b | 16.1% | ||||
| 11.2% | |||||
| gpt-oss-20b | 9.80% | ||||
| 4.90% |
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.
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.
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
- 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.