E-Bench (Tencent)
Agents complete Tencent in-house work tasks in the working agent group. Task and scoring details are not published.
Not the Shanghai AI Lab robotics test EBench. Hy3 scored 50.2 in its own card and 48.5 in the Hy4 preview card after harness updates.
- Made by
- Tencent Hy team
- Released
- July 2026
- Kind
- Documents and spreadsheets
- Level
- Expert
- Scored
- Checked automatically
- Answers
- Agent task
- Tasks
- Kept private
- Strongest model tested
- Claude Opus 5 #10 on the capability index
- Last in a launch post
- August 2026
- Last model added
- August 2026
- Scores refreshed
- November 12th, 2026
GPT-5.6 Sol leads with 80.6% among models released by August 2026
Best score per model from the benchmark's own leaderboard, read on November 12th, 2026. Where a model ran with several settings or harnesses, we keep its best.
GPT-5.6 Sol leads with 80.6%, 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.
The newest model with a score here came out in August 2026, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:
- Claude Fable 5.2
-
Gemini 4 Carbon - GPT-6 Astra
- Claude Opus 5.5
- and 14 more
- 1 GPT-5.6 Sol highest available reasoning 80.6%
- 2 Claude Opus 5 highest available reasoning 77.8%
- 3
Tencent Hy4 preview highest available reasoning 77.1% - 4 Kimi K3 highest available reasoning 73.8%
- 5 GLM-5.3 highest available reasoning 71.4%
- 6
Qwen 3.8 Max highest available reasoning 66.8% - 7
DeepSeek V4 Pro 0813 highest available reasoning 61.3% - 8 Tencent Hy3 highest available reasoning 48.5%
Source: Official E-Bench (Tencent) leaderboard
- 1 GPT-5.6 Sol highest available reasoning 80.6%
- 2 Claude Opus 5 highest available reasoning 77.8%
- 3 Tencent Hy4 preview highest available reasoning 77.1%
- 4 Kimi K3 highest available reasoning 73.8%
- 5 GLM-5.3 highest available reasoning 71.4%
- 6 Qwen 3.8 Max highest available reasoning 66.8%
- 7 DeepSeek V4 Pro 0813 highest available reasoning 61.3%
- 8 Tencent Hy3 50.2%
Sources: Official E-Bench (Tencent) leaderboard and labs’ model cards
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | GPT-5.6 Sol | 80.6% | highest available reasoning | Jul 2026 | Official leaderboard |
| 2 | Claude Opus 5 | 77.8% | highest available reasoning | Jul 2026 | Official leaderboard |
| 3 | Tencent Hy4 preview | 77.1% | highest available reasoning | Aug 2026 | Official leaderboard |
| 4 | Kimi K3 | 73.8% | highest available reasoning | Jul 2026 | Official leaderboard |
| 5 | GLM-5.3 | 71.4% | highest available reasoning | Aug 2026 | Official leaderboard |
| 6 | Qwen 3.8 Max | 66.8% | highest available reasoning | Aug 2026 | Official leaderboard |
| 7 | DeepSeek V4 Pro 0813 | 61.3% | highest available reasoning | Aug 2026 | Official leaderboard |
| 8 | Tencent Hy3 | 48.5% | highest available reasoning | Jul 2026 | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | GPT-5.6 Sol | 80.6% | highest available reasoning | Jul 2026 | Official leaderboard |
| 2 | Claude Opus 5 | 77.8% | highest available reasoning | Jul 2026 | Official leaderboard |
| 3 | Tencent Hy4 preview | 77.1% | highest available reasoning | Aug 2026 | Official leaderboard |
| 4 | Kimi K3 | 73.8% | highest available reasoning | Jul 2026 | Official leaderboard |
| 5 | GLM-5.3 | 71.4% | highest available reasoning | Aug 2026 | Official leaderboard |
| 6 | Qwen 3.8 Max | 66.8% | highest available reasoning | Aug 2026 | Official leaderboard |
| 7 | DeepSeek V4 Pro 0813 | 61.3% | highest available reasoning | Aug 2026 | Official leaderboard |
| 8 | Tencent Hy3 | 50.2% | Jul 2026 | Lab model card |
Labs report scores for 2 models in their cards. Switch the scores above to rank them together.
2 models have held number one on the official leaderboard since its release in July 2026 2 models have held number one with claims included since its release in July 2026
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official E-Bench (Tencent) 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 E-Bench (Tencent) leaderboard and labs’ model cards
One lab put it in a launch post in the last six months
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 |
|---|---|---|---|
| Aug 28, 2026 | Tencent Hy4 preview | 77.1 | E-Bench (Internal) |
| Jul 2, 2026 | Tencent Hy3 | 50.2 | e-bench (internal) |
2 lab claims next to the leaderboard
Each score a lab reported for its own model, beside the same model's best score on the leaderboard. Gaps mostly come from different harnesses, effort levels and task subsets. 1 of them match, which usually means one side copied the other. Every benchmark.
| Model | Lab says | Board says | Gap |
|---|---|---|---|
| Tencent Hy3 | 50.2% e-bench (internal) | 48.5% Leaderboard, highest available reasoning | +1.7 pts |
| Tencent Hy4 preview | 77.1% E-Bench (Internal) | 77.1% Leaderboard, highest available reasoning | Same |
More in Economic work
- JobBench In use 130 tasks from 35 occupations workers most want delegated. Agent works through messy reference files, graded by fact-anchored binary rubrics. GLM-5.3 61.4%
- GDPval In use Real work tasks from 44 occupations with reference files. Experts blindly compare model deliverables with expert work. Metric is win-or-tie rate.
Seed2.1 Pro 87.9% - GDPval-AA v2.1 In use Agents with shell and web access complete 220 GDPval gold tasks from 44 occupations. Blind pairwise comparisons give Elo ratings. Claude Opus 5.5 1,846
- AA-Briefcase v1.1 In use Version 1.1 of Artificial Analysis's multi-hour office work evaluation, with Elo ratings fitted by Crowd-BT. Claude Opus 5.5 1,822
- CoWorkBench In use Agent completes long-horizon productivity tasks across computer science, finance, law and medicine. Qwen in-house scoring (method unpublished). Claude Fable 5 75.9%
- OfficeQA In use Grounded reasoning over a large corpus of U.S. Treasury Bulletin documents: locate tables and compute precise numeric answers. Claude Opus 4.7 86.3%