AI Gaming Passport · VG-Agent (VideoGameBench reference agent)
VG-Agent (VideoGameBench reference agent)
2 GAMESBenchmark reference agent · project: VideoGameBench · games: Kirby's Dream Land, Pokémon Crystal · creator profile
Atom feed for VG-Agent (VideoGameBench reference agent)
Follow saves this passport in your browser; changes show on the home page. No account or cookies. The feed works in any feed reader.
Passport card
- Name
- VG-Agent (VideoGameBench reference agent)
- Type
- Benchmark reference agent
- Maker or creator
- VideoGameBench
- Status
- Published (benchmark result)
- Models used
- GPT-4o, Gemini 2.5 Pro (preview 03-25)
- Games played
- Kirby's Dream Land, Pokémon Crystal
- First seen
- Oct 18, 2026
- Last verified
- Oct 18, 2026
- Last checked
- Oct 18, 2026
- Evidence by label
- Verified 2 records, 4 source links
Creator-reported 0 records, 0 source links
Awaiting verification 0 records, 0 source links
Best verified result per game
One row per game. Results in different games are never combined or compared.
| Game | Best verified result | How picked | Last verified |
|---|---|---|---|
| Kirby's Dream Land | No verified result yet | – | Unknown |
| Pokémon Crystal | No verified result yet | – | Unknown |
How to cite: AI vs Games, “VG-Agent (VideoGameBench reference agent): AI Gaming Passport,” https://aivsgames.com/agents/vg-agent/ (last checked Oct 18, 2026). Please also cite the original sources linked on each record.
Achievements
No sourced achievements yet.
Memorable failures
| Model | Game | Achievement | Status | Date |
|---|---|---|---|---|
| Gemini 2.5 Pro (preview 03-25) | Kirby's Dream Land | Memorable fail 4.8% of Kirby's Dream Land (best single-game score on the VideoGameBench test set; 0.48% overall) Last verified: Oct 18, 2026 · Last checked: Oct 18, 2026 How was this run measured?
|
Verified | May 23, 2025 |
| GPT-4o (2024-08-06) | Pokémon Crystal | Memorable fail 0.9% of Pokémon Crystal completed on VideoGameBench (test set, real-time) Picked Cyndaquil as its starter, then lost track of what it was supposed to do. Last verified: Oct 18, 2026 · Last checked: Oct 18, 2026 How was this run measured?
|
Verified | May 23, 2025 |
Runs
| Run | Game | Models (role) | Status | Started |
|---|---|---|---|---|
| vgb-gemini-2-5-pro-kirby | Kirby's Dream Land | Gemini 2.5 Pro (preview 03-25) (player) | published | – |
| vgb-gpt-4o-crystal | Pokémon Crystal | GPT-4o (player) | published | – |
Model versions used
Milestones timeline
- May 23, 2025
✗ Kirby's Dream Land · Gemini 2.5 Pro (preview 03-25): 4.8% of Kirby's Dream Land (best single-game score on the VideoGameBench test set; 0.48% overall) Verified - May 23, 2025
✗ Pokémon Crystal · GPT-4o (2024-08-06): 0.9% of Pokémon Crystal completed on VideoGameBench (test set, real-time) Verified
This Week in AI vs Games
A short weekly email with new records, memorable failures and runs to watch, every claim linked to its source.
Newsletter launching soon. Follow passports or subscribe by feed for now: site-wide Atom feed · passports to follow.
Latest issue: Issue #1: The Ender Dragon is down