Skip to main content

ResearchClawBench

An end-to-end autonomous scientific research benchmark with 40 tasks across 10 scientific domains, where agents receive related literature and raw data, then attempt to rediscover the hidden target paper.

How BenchLM shows ResearchClawBench

BenchLM mirrors the official ResearchClawBench Pass@1 leaderboard snapshot. The source benchmark contains 40 tasks across 10 scientific domains and uses RADS average (Pass@1) as the primary metric.

ResearchClawBench gives agents related literature and raw data, hides the target paper, and grades how much of the scientific result they rediscover. The RADS scale treats 50 as matching the original paper and 70+ as surpassing it.

ResearchClawBench is display only on BenchLM. The rows combine a model, a research harness, execution budget, and long-horizon scientific workflow, so BenchLM does not use them as weighted base-model ranking inputs.

28 agent rows20 mapped ResearchHarness rows40 tasks10 domains6 Pass@5 rows trackedDisplay only

RADS average (Pass@1) on ResearchClawBench — 2026-06-29 snapshot

BenchLM mirrors the published rads average (pass@1) view for ResearchClawBench. Claude Code leads the public snapshot at 21.5% , followed by Claude Opus 4.8 (21.1%) and Claude Opus 4.7 (20.7%). BenchLM does not use these results to rank models overall.

28 modelsAgenticCurrentDisplay onlyUpdated 2026-06-29 snapshot

The published ResearchClawBench snapshot is tightly clustered at the top: Claude Code sits at 21.5%, while the third row is only 0.8 points behind. The broader top-10 spread is 3.3 points, so many of the published scores sit in a relatively narrow band.

28 models have been evaluated on ResearchClawBench. The benchmark falls in the Agentic category. This category carries a 22% weight in BenchLM.ai's overall scoring system. ResearchClawBench is currently displayed for reference but excluded from the scoring formula, so it does not directly affect overall rankings.

About ResearchClawBench

Year

2026

Tasks

40 tasks across 10 scientific domains

Format

End-to-end autonomous research evaluation with RADS scoring

Difficulty

Scientific research re-discovery

ResearchClawBench grades scientific agents with RADS, a rubric where 50 indicates matching the target paper and 70+ indicates surpassing it. BenchLM mirrors the official Pass@1 leaderboard as display-only because rows reflect a research-agent harness and long-horizon scientific workflow, not a normalized base-model-only comparison.

BenchLM freshness & provenance

Version

ResearchClawBench 2026

Refresh cadence

Quarterly

Staleness state

Current

Question availability

Public benchmark set

CurrentDisplay only

BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.

RADS average (Pass@1) table (28 models)

1
Claude Codeclaude-opus-4-6
21.5%
2
Claude Opus 4.8claude-opus-4-8
21.1%
3
Claude Opus 4.7Claude-Opus-4.7
20.7%
4
20.7%
5
Claude Opus 4.6Claude-Opus-4.6
19.9%
6
MiniMax M3MiniMax-M3
19.8%
7
18.8%
8
Qwen3.7 Maxqwen3.7-max
18.7%
9
Codex CLIgpt-5.4
18.4%
10
18.2%
11
Kimi K2.6Kimi-K2.6
18.0%
12
Qwen3.6 PlusQwen3.6-Plus
18.0%
13
Gemini 3.5 Flashgemini-3.5-flash
17.9%
14
DeepSeek V4 ProDeepSeek-V4-Pro
17.1%
15
17.0%
16
16.9%
17
OpenClawgpt-5.4
16.6%
18
16.3%
19
15.5%
20
15.3%
21
15.3%
22
Qwen3.5 397BQwen3.5-397B-A17B
14.2%
23
Kimi K2.5Kimi-K2.5
14.0%
24
ARIS Codexgpt-5.4
13.6%
25
Grok 4.1Grok-4.1
13.5%
26
Gemini 3.1 ProGemini-3.1-Pro
13.3%
27
Nanobotgpt-5.4
12.8%
28
Grok 4.3Grok-4.3
12.4%

FAQ

What does ResearchClawBench measure?

An end-to-end autonomous scientific research benchmark with 40 tasks across 10 scientific domains, where agents receive related literature and raw data, then attempt to rediscover the hidden target paper.

Which model leads the published ResearchClawBench snapshot?

Claude Code currently leads the published ResearchClawBench snapshot with 21.5% rads average (pass@1). BenchLM shows this benchmark for display only and does not use it in overall rankings.

How many models are evaluated on ResearchClawBench?

28 AI models are included in BenchLM's mirrored ResearchClawBench snapshot, based on the public leaderboard captured on 2026-06-29 snapshot.

Last updated: 2026-06-29 snapshot · mirrored from the public benchmark leaderboard

The AI models change fast. We track them for you.

For engineers, researchers, and the plain curious — a weekly brief on new models, ranking shifts, and pricing changes.

Free. No spam. Unsubscribe anytime.