Research-Eval
Gemini 2.0 Flash Feb 2025 54.2% leads since August 2025 Search-augmented models answer 374 grounded multi-hop questions that need sources across many kinds of websites. An LLM judge checks each answer against a checklist of requirements.
Reka in-house tutors wrote 606 candidate questions and kept 374 after double annotation. Reka recommends deepseek-v3-0324 as the judge so results stay comparable.
- Made by
- Reka AI
- Released
- August 2025
- Kind
- Deep research
- Level
- Expert
- Scored
- Graded by a model
- Answers
- Short answer
- Tasks
- All public
- Strongest model tested
- Grok 4 #85 on the capability index
- Last in a launch post
- November 2025, dropped before it saturated
- Last model added
- November 2025
- Scores refreshed
- November 12th, 2026
Grok 4 Fast leads with 66%
Best score per model as labs report them in their own cards. Labs pick their own settings, so these compare less cleanly than independent runs.
Gemini 2.0 Flash leads with 54.2% across all independent runs
Best score per model across the official leaderboard, Epoch AI’s own runs and any board Epoch AI copied. Each bar and row names its source.
Grok 4 Fast leads with 66%, claims included
Best score per model from every source, including labs’ own numbers and claims Epoch AI collected from reports and papers. Labs pick their own settings, so a claim can beat an independent run of the same model.
- 1 Grok 4 Fast 66%
- 2 Grok 4.1 Fast 63.9%
Source: Labs’ model cards
- 1 Gemini 2.0 Flash Feb 2025 54.2%
- 2 Claude Sonnet 4 44.8%
- 3 Grok 4 26.7%
Source: Official Research-Eval leaderboard
- 1 Grok 4 Fast 66%
- 2 Grok 4.1 Fast 63.9%
- 3 Gemini 2.0 Flash Feb 2025 54.2%
- 4 Claude Sonnet 4 44.8%
- 5 Grok 4 26.7%
Sources: Labs’ model cards and official Research-Eval leaderboard
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Grok 4 Fast | 66% | Sep 2025 | Lab model card | |
| 2 | Grok 4.1 Fast | 63.9% | Nov 2025 | Lab model card |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Gemini 2.0 Flash Feb 2025 | 54.2% | Google Search | Feb 2025 | Official leaderboard |
| 2 | Claude Sonnet 4 | 44.8% | web search | May 2025 | Official leaderboard |
| 3 | Grok 4 | 26.7% | live search | Jul 2025 | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Grok 4 Fast | 66% | Sep 2025 | Lab model card | |
| 2 | Grok 4.1 Fast | 63.9% | Nov 2025 | Lab model card | |
| 3 | Gemini 2.0 Flash Feb 2025 | 54.2% | Google Search | Feb 2025 | Official leaderboard |
| 4 | Claude Sonnet 4 | 44.8% | web search | May 2025 | Official leaderboard |
| 5 | Grok 4 | 26.7% | live search | Jul 2025 | Official leaderboard |
The benchmark's own leaderboard has 7 entries, 3 of them models we track. Switch the scores above to rank them together.
Number one over time Number one over time 2 models have held number one with claims included since its release in August 2025
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official Research-Eval leaderboard
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Sources: Official Research-Eval leaderboard and labs’ model cards
No launch post has listed it since November 2025
Launch posts from every lab that has listed it, one tick per post. Tall colored ticks include it, and short gray ones leave it out.
| Date | Launch post | Score | As the post writes it |
|---|---|---|---|
| Nov 19, 2025 | Grok 4.1 Fast | 63.9 | Research-Eval Reka (Grok 4.1 Fast, Agent Tools API) |
| Sep 19, 2025 | Grok 4 Fast | 66 | Reka Research Eval pass@1 |
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 Carbon 53.9%
- 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%