GameASG-Bench
Agents build 47 browser games from written requirements. Static checks inspect each HTML file, and browser tests check controls, gameplay and state changes.
The September 18, 2026 paper compares nine model and harness combinations. GPT-6-Astra with Codex CLI led with 26 of 47 games passing (55.3%).
- Made by
- AReaL project
- Released
- September 2026
- Kind
- Building whole projects
- Level
- University
- Scored
- Checked automatically
- Answers
- Code
- Tasks
- All public
- Measures
- Capability
- Language
- English
- Interaction
- Agentic
- Refresh
- One fixed set
- Run by
- Anyone
- In views
- Agents
- Strongest model tested
- GPT-6 Astra #4 on the capability index
- Last model added
- September 2026
- Scores refreshed
- November 20th, 2026
GPT-6 Astra leads with 55.3% among models released by September 2026 GPT-6 Astra leads with 55.3% among models released by September 2026
Best score per entry from the benchmark's paper, read on November 20th, 2026. A paper’s table doesn’t add newer models. Where a model ran with several settings or harnesses, we keep its best. Every run is in the next chart. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. This view shows only models we track, each at its best.
The newest model with a score here came out in September 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
- Claude Opus 5.5
- and 13 more
- 1 GPT-6 Astra Maximum reasoning effort 55.3%
- 2 Claude Opus 5 Maximum reasoning effort 51.1%
- 3 GPT-5.6 Sol Maximum reasoning effort 44.7%
- 4
DeepSeek-V4-Flash High reasoning effort 40.4%
- 5
DeepSeek-V4-Pro Maximum reasoning effort 31.9%
- 6 Kimi K3 Maximum reasoning effort 31.9%
- 7 GLM-5.2 Maximum reasoning effort 23.4%
- 8 Hunyuan-3 Maximum reasoning effort 21.3%
- 9
MiniMax-M3 Maximum reasoning effort 14.9%
Source: The GameASG-Bench paper
- 1 GPT-6 Astra Maximum reasoning effort 55.3%
- 2 Claude Opus 5 Maximum reasoning effort 51.1%
- 3 GPT-5.6 Sol Maximum reasoning effort 44.7%
- 4
DeepSeek-V4-Flash High reasoning effort 40.4%
- 5
DeepSeek-V4-Pro Maximum reasoning effort 31.9%
- 6 Kimi K3 Maximum reasoning effort 31.9%
- 7 GLM-5.2 Maximum reasoning effort 23.4%
- 8
MiniMax-M3 Maximum reasoning effort 14.9%
Source: The GameASG-Bench paper
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | GPT-6 Astra | 55.3% | Maximum reasoning effort, Codex CLI | Sep 2026 | Official leaderboard |
| 2 | Claude Opus 5 | 51.1% | Maximum reasoning effort, Claude Code | Jul 2026 | Official leaderboard |
| 3 | GPT-5.6 Sol | 44.7% | Maximum reasoning effort, Codex CLI | Jul 2026 | Official leaderboard |
| 4 | | 40.4% | High reasoning effort, Claude Code | Apr 2026 | Official leaderboard |
| 5 | | 31.9% | Maximum reasoning effort, Claude Code | Apr 2026 | Official leaderboard |
| 6 | Kimi K3 | 31.9% | Maximum reasoning effort, Claude Code | Jul 2026 | Official leaderboard |
| 7 | GLM-5.2 | 23.4% | Maximum reasoning effort, Claude Code | Jun 2026 | Official leaderboard |
| 8 | Hunyuan-3 | 21.3% | Maximum reasoning effort, Claude Code | Official leaderboard | |
| 9 | | 14.9% | Maximum reasoning effort, Claude Code | Jun 2026 | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | GPT-6 Astra | 55.3% | Maximum reasoning effort, Codex CLI | Sep 2026 | Official leaderboard |
| 2 | Claude Opus 5 | 51.1% | Maximum reasoning effort, Claude Code | Jul 2026 | Official leaderboard |
| 3 | GPT-5.6 Sol | 44.7% | Maximum reasoning effort, Codex CLI | Jul 2026 | Official leaderboard |
| 4 | | 40.4% | High reasoning effort, Claude Code | Apr 2026 | Official leaderboard |
| 5 | | 31.9% | Maximum reasoning effort, Claude Code | Apr 2026 | Official leaderboard |
| 6 | Kimi K3 | 31.9% | Maximum reasoning effort, Claude Code | Jul 2026 | Official leaderboard |
| 7 | GLM-5.2 | 23.4% | Maximum reasoning effort, Claude Code | Jun 2026 | Official leaderboard |
| 8 | | 14.9% | Maximum reasoning effort, Claude Code | Jun 2026 | Official leaderboard |
Rankings by part
We read 5 parts from the benchmark’s paper, each ranking models on one part of the test, such as L1, L2 and L2 P0. DeepSeek-V4-Flash leads 4 of them, and GPT-6 Astra leads the other one. Arrows show how far a model moved from its overall rank among the models we track.
L1
8 models we track have a score here.
- 1
DeepSeek-V4-Flash ↑3 100%
- 2
DeepSeek-V4-Pro ↑3 99.6%
- 3 Claude Opus 5 ↓1 99.6%
- 4 GPT-5.6 Sol ↓1 98.9%
- 5 GLM-5.2 ↑2 98.8%
- 6
MiniMax-M3 ↑2 98.3%
- 7 GPT-6 Astra ↓6 98%
- 8 Kimi K3 ↓2 97.7%
L2
8 models we track have a score here.
- 1 GPT-6 Astra 93.2%
- 2 GPT-5.6 Sol ↑1 91.3%
- 3 Claude Opus 5 ↓1 90.4%
- 4
DeepSeek-V4-Flash 89.3%
- 5 Kimi K3 ↑1 88.5%
- 6
DeepSeek-V4-Pro ↓1 85.2%
- 7 GLM-5.2 84.6%
- 8
MiniMax-M3 70.6%
L2 P0
8 models we track have a score here.
- 1
DeepSeek-V4-Flash ↑3 100%
- 2 GPT-5.6 Sol ↑1 100%
- 3 GPT-6 Astra ↓2 99%
- 4 Kimi K3 ↑2 98%
- 5
MiniMax-M3 ↑3 91.2%
- 6 GLM-5.2 ↑1 91.2%
- 7 Claude Opus 5 ↓5 91.2%
- 8
DeepSeek-V4-Pro ↓3 85.3%
L2 P1
8 models we track have a score here.
- 1
DeepSeek-V4-Flash ↑3 94%
- 2 GPT-6 Astra ↓1 92.7%
- 3 Claude Opus 5 ↓1 89.9%
- 4 GPT-5.6 Sol ↓1 89.7%
- 5 Kimi K3 ↑1 86.3%
- 6
DeepSeek-V4-Pro ↓1 83.9%
- 7 GLM-5.2 83.3%
- 8
MiniMax-M3 64.5%
L2 P2
8 models we track have a score here.
- 1
DeepSeek-V4-Flash ↑3 97.6%
- 2 GPT-6 Astra ↓1 92.4%
- 3 GPT-5.6 Sol 91.9%
- 4 Claude Opus 5 ↓2 91.1%
- 5 Kimi K3 ↑1 89.7%
- 6
DeepSeek-V4-Pro ↓1 88.9%
- 7 GLM-5.2 83%
- 8
MiniMax-M3 75.4%
The best score, 55.3%, costs $4.44 a task
Runs on the cost frontier score higher than any run that's cheaper. The costs are the leaderboard's own estimates, so compare them only within this benchmark.
Source: The GameASG-Bench paper
| Model | Effort | Harness | Score | Cost per task | Output tokens |
|---|---|---|---|---|---|
| GPT-6 Astra | Maximum reasoning effort | Codex CLI | 55.3% | $4.44 | 55,500 |
| Claude Opus 5 | Maximum reasoning effort | Claude Code | 51.1% | $12 | 142,400 |
| GPT-5.6 Sol | Maximum reasoning effort | Codex CLI | 44.7% | $1.79 | 45,700 |
| High reasoning effort | Claude Code | 40.4% | |||
| Maximum reasoning effort | Claude Code | 38.3% | $0.25 | 203,600 | |
| Maximum reasoning effort | Codex CLI | 38.3% | |||
| Maximum reasoning effort | Claude Code | 31.9% | $0.74 | 202,900 | |
| Kimi K3 | Maximum reasoning effort | Claude Code | 31.9% | $2.55 | 72,300 |
| 60 turns | Claude Code | 27.7% | |||
| GLM-5.2 | Maximum reasoning effort | Claude Code | 23.4% | $0.62 | 47,000 |
| Files + syntax checking | Claude Code | 19.1% | |||
| No tools (serialized API) | Serialized API | 14.9% | |||
| 30 turns | Claude Code | 14.9% | |||
| Low reasoning effort | Claude Code | 14.9% | |||
| Maximum reasoning effort | Claude Code | 14.9% | $0.41 | 71,100 | |
| File read/write only | Claude Code | 12.8% |
The best score, 55.3%, costs $4.44 a task
Runs on the cost frontier score higher than any run that's cheaper. The costs are the leaderboard's own estimates, so compare them only within this benchmark.
Source: The GameASG-Bench paper
| Model | Effort | Harness | Score | Cost per task | Output tokens |
|---|---|---|---|---|---|
| GPT-6 Astra | Maximum reasoning effort | Codex CLI | 55.3% | $4.44 | 55,500 |
| Claude Opus 5 | Maximum reasoning effort | Claude Code | 51.1% | $12 | 142,400 |
| GPT-5.6 Sol | Maximum reasoning effort | Codex CLI | 44.7% | $1.79 | 45,700 |
| High reasoning effort | Claude Code | 40.4% | |||
| Maximum reasoning effort | Claude Code | 38.3% | $0.25 | 203,600 | |
| Maximum reasoning effort | Codex CLI | 38.3% | |||
| Maximum reasoning effort | Claude Code | 31.9% | $0.74 | 202,900 | |
| Kimi K3 | Maximum reasoning effort | Claude Code | 31.9% | $2.55 | 72,300 |
| 60 turns | Claude Code | 27.7% | |||
| GLM-5.2 | Maximum reasoning effort | Claude Code | 23.4% | $0.62 | 47,000 |
| Files + syntax checking | Claude Code | 19.1% | |||
| No tools (serialized API) | Serialized API | 14.9% | |||
| 30 turns | Claude Code | 14.9% | |||
| Low reasoning effort | Claude Code | 14.9% | |||
| Maximum reasoning effort | Claude Code | 14.9% | $0.41 | 71,100 | |
| File read/write only | Claude Code | 12.8% |
2 models have held number one in the benchmark’s paper since its release in September 2026
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: The GameASG-Bench paper
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official GameASG-Bench 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 Coding
- SWE-bench Verified In use 500 human-validated GitHub issues from 12 Python repos. Agent writes a patch that must pass hidden tests. Claude Opus 4.7 83.5%
- LiveCodeBench v6 In use Competitive programming problems collected after model cutoffs from LeetCode, AtCoder, Codeforces. Scored pass@1 on hidden tests.
DeepSeek-V4-Pro 93.5%
- SWE-bench Multilingual In use 300 GitHub-issue tasks across nine programming languages in SWE-bench format. Patch graded by repository tests. Claude Fable 5.2 96.6%
- DeepSWE In use 113 original long-horizon engineering tasks across 91 repos and five languages, written from scratch. Hand-written verifiers accept alternative solutions. GPT-6 Astra 74.1%
- SciCode In use Scientist-curated research coding problems split into subproblems across natural sciences. Code checked against test cases.
Seed2.1 Pro 59.8%
- Codeforces rating In use Model solves recent Codeforces contest problems. An Elo-style rating is computed from simulated contest performance.
DeepSeek V4.1 Flash 3,471
Data audit
All 10 checks pass. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 4 pass 1 note
- Pass Independent runs Scores from the benchmark’s paper (9 entries).
- Pass Models scored 8 models we track, plus 1 other entry on its leaderboard.
- Pass Newest score The newest is from September 2026, for GPT-6 Astra.
- Note Frontier models Only 1 of the top 10 has a score. The strongest tested is GPT-6 Astra (#4).
- Pass Headroom The best is 55.3%, under 85%, so it still separates models.
Leaderboard import 3 pass
- Pass Importer We read the board on arxiv.org: 16 rows for models we track and 1 other entry 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.