QwenWorldBench
Models act as world models that simulate agent environments in 7 domains, such as terminal, web, and Android. A rubric judge scores five dimensions.
In-house: Qwen never released the tasks.
- Made by
- Qwen Team, Alibaba Group
- Released
- May 2026
- Kind
- Interactive world models
- Scored
- Graded by a model
- Answers
- Open-ended
- Tasks
- Kept private
- Strongest model tested
- Claude Opus 4.6 #37 on the capability index
- Last in a launch post
- June 2026, dropped before it saturated
- Last model added
- June 2026
- Scores refreshed
- November 12th, 2026
Qwen3.7-Plus leads with 62.1% among models released by June 2026
Best score per entry from the benchmark's own leaderboard, read on November 12th, 2026. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders.
Qwen3.7-Plus leads with 62.1%, 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 June 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 16 more
- 1
Qwen3.7-Plus 62.1% - 2 Claude Opus 4.6 max 56.1%
- 3 DeepSeek V4 Pro (snapshot unspecified) max 52.3%
- 4 Kimi K2.6 thinking 50.9%
- 5 GLM-5.1 thinking 50.2%
- 6 Qwen 3.6 Plus 47.6%
Source: Official QwenWorldBench leaderboard
- 1 Qwen3.7-Plus 62.1%
- 2 Qwen3.7-Max 57.3%
- 3 Claude Opus 4.6 max 56.1%
- 4 Kimi K2.6 thinking 50.9%
- 5 GLM-5.1 thinking 50.2%
- 6 Qwen 3.6 Plus 47.6%
Sources: Official QwenWorldBench leaderboard and labs’ model cards
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Qwen3.7-Plus | 62.1% | Jun 2026 | Official leaderboard | |
| 2 | Claude Opus 4.6 | 56.1% | max | Feb 2026 | Official leaderboard |
| 3 | DeepSeek V4 Pro (snapshot unspecified) | 52.3% | max | Official leaderboard | |
| 4 | Kimi K2.6 | 50.9% | thinking | Apr 2026 | Official leaderboard |
| 5 | GLM-5.1 | 50.2% | thinking | Apr 2026 | Official leaderboard |
| 6 | Qwen 3.6 Plus | 47.6% | Mar 2026 | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Qwen3.7-Plus | 62.1% | Jun 2026 | Official leaderboard | |
| 2 | Qwen3.7-Max | 57.3% | May 2026 | Lab model card | |
| 3 | Claude Opus 4.6 | 56.1% | max | Feb 2026 | Official leaderboard |
| 4 | Kimi K2.6 | 50.9% | thinking | Apr 2026 | Official leaderboard |
| 5 | GLM-5.1 | 50.2% | thinking | Apr 2026 | Official leaderboard |
| 6 | Qwen 3.6 Plus | 47.6% | Mar 2026 | Official leaderboard |
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 May 2026 3 models have held number one with claims included since its release in May 2026
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official QwenWorldBench 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 QwenWorldBench leaderboard and labs’ model cards
Labs headline it less often now, last in June 2026
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 |
|---|---|---|---|
| Jun 1, 2026 | Qwen3.7-Plus | 62.1 | QwenWorldBench |
| May 20, 2026 | Qwen3.7-Max | 57.3 | QwenWorldBench |
One lab claim 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 |
|---|---|---|---|
| Qwen3.7-Plus | 62.1% QwenWorldBench | 62.1% Leaderboard | Same |
More in World models
- AgentWorldBench In use Language world models predict the next environment observation after an agent action, in 7 domains from MCP and search to Android and desktop OS. A judge model scores each prediction against the real observation on five dimensions. GPT-5.4 58.3
- IntPhys 2 In use Models watch simulated videos and judge whether each event is physically possible, testing object permanence, immutability, continuity and solidity. Scored by accuracy.
Muse Spark Open 68.3% - VPCT Fading Models view an image of a ball, ramps, and three buckets and predict which bucket the ball falls into. Scored by accuracy on 100 problems. Gemini 3 Pro 91%
- ContPhy Fading Models infer physical properties such as mass and density of fluids and soft bodies from videos and predict their dynamics.
Seed2.0 Pro 67.4% - RBench Only in system cards 650 image+text prompts for robot video generation (tasks and embodiments). MLLM judge scores task completion, physical plausibility, stability.
Wan 2.6 60.7% - PAI-Bench Only in system cards Physical-AI video generation from 1,044 prompts in six domains, VLM-QA plus quality metrics. Companion control and understanding tracks.
Cosmos3-Nano 83.7%