Crest Arcade Starjar
Models place pieces in a jar so matching pieces touch and merge. The score compares their total points with a reference bot on identical jars.
Glimmer Index publishes scores for 15 distinct models. Of those, 10 completed the required set.
- Made by
- Glimmer Index
- Kind
- Video games
- Level
- Everyday
- Scored
- Checked automatically
- Answers
- Agent task
- Tasks
- All public
- Measures
- Capability
- Language
- English
- Interaction
- Agentic
- Environment
- Starjar
- Items from
- Generated
- Refresh
- One fixed set
- Run by
- Anyone
- Strongest model tested
- GPT-6.2 Astra #1 on the capability index
- Last model added
- November 2026
- Scores refreshed
- November 20th, 2026
Gemini 4 Neon leads with 114.5 Gemini 4 Neon leads with 114.5
Best score per entry from the benchmark's own leaderboard, read on November 20th, 2026. 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.
- 1
Gemini 4 Neon medium · Provisional 114.5
- 2 GPT-6 Astra medium 112.3
- 3
Qwen 3.8 27B none · Provisional 108.1
- 4 Claude Sonnet 5.5 medium 105.7
- 5 GPT-6.1 Terra medium 103.7
- 6
Qwen3.8-Flash-Next low 103.3
- 7
Gemini 4 Hydrogen medium 102.9
- 8 GPT-6.1 Sol medium 101.6
- 9 Claude Haiku 5.5 medium 98.7
- 10 GPT-6.2 Astra medium 98.7
- 11
Gemma 3 4B none · Provisional 98.2
- 12 GPT-6 Luna medium 95.2
- 13 Ministral 3 8B none · Provisional 93.0
- 14 GPT-6 Sol medium 90.0
- 15
Qwen3.5-9B none · Provisional 0.0
Source: Official Crest Arcade Starjar leaderboard
- 1
Gemini 4 Neon medium · Provisional 114.5
- 2 GPT-6 Astra medium 112.3
- 3
Qwen 3.8 27B none · Provisional 108.1
- 4 Claude Sonnet 5.5 medium 105.7
- 5 GPT-6.1 Terra medium 103.7
- 6
Qwen3.8-Flash-Next low 103.3
- 7
Gemini 4 Hydrogen medium 102.9
- 8 GPT-6.1 Sol medium 101.6
- 9 Claude Haiku 5.5 medium 98.7
- 10 GPT-6.2 Astra medium 98.7
- 11
Gemma 3 4B none · Provisional 98.2
- 12 GPT-6 Luna medium 95.2
- 13 GPT-6 Sol medium 90.0
- 14
Qwen3.5-9B none · Provisional 0.0
Source: Official Crest Arcade Starjar leaderboard
| # | Entry | Score ± | Setting | Released | Run on | Source |
|---|---|---|---|---|---|---|
| 1 | | 114.5 ±17.6 | medium · Provisional | Nov 2026 | Nov 2026 | Official leaderboard |
| 2 | GPT-6 Astra | 112.3 ±17.3 | medium | Sep 2026 | Oct 2026 | Official leaderboard |
| 3 | | 108.1 ±12.1 | none · Provisional | Aug 2026 | Nov 2026 | Official leaderboard |
| 4 | Claude Sonnet 5.5 | 105.7 ±5.6 | medium | Sep 2026 | Nov 2026 | Official leaderboard |
| 5 | GPT-6.1 Terra | 103.7 ±4.8 | medium | Oct 2026 | Oct 2026 | Official leaderboard |
| 6 | | 103.3 ±12 | low | Aug 2026 | Nov 2026 | Official leaderboard |
| 7 | | 102.9 ±6.9 | medium | Nov 2026 | Nov 2026 | Official leaderboard |
| 8 | GPT-6.1 Sol | 101.6 ±13.4 | medium | Sep 2026 | Oct 2026 | Official leaderboard |
| 9 | Claude Haiku 5.5 | 98.7 ±11.7 | medium | Oct 2026 | Nov 2026 | Official leaderboard |
| 10 | GPT-6.2 Astra | 98.7 ±14 | medium | Nov 2026 | Nov 2026 | Official leaderboard |
| 11 | | 98.2 ±37.9 | none · Provisional | Mar 2025 | Nov 2026 | Official leaderboard |
| 12 | GPT-6 Luna | 95.2 ±4.4 | medium | Sep 2026 | Oct 2026 | Official leaderboard |
| 13 | Ministral 3 8B open weights | 93.0 ±8.8 | none · Provisional | Nov 2026 | Official leaderboard | |
| 14 | GPT-6 Sol | 90.0 ±10.5 | medium | Sep 2026 | Oct 2026 | Official leaderboard |
| 15 | | 0.0 0 to 0.1 | none · Provisional | Feb 2026 | Nov 2026 | Official leaderboard |
| # | Model | Score ± | Setting | Released | Run on | Source |
|---|---|---|---|---|---|---|
| 1 | | 114.5 ±17.6 | medium · Provisional | Nov 2026 | Nov 2026 | Official leaderboard |
| 2 | GPT-6 Astra | 112.3 ±17.3 | medium | Sep 2026 | Oct 2026 | Official leaderboard |
| 3 | | 108.1 ±12.1 | none · Provisional | Aug 2026 | Nov 2026 | Official leaderboard |
| 4 | Claude Sonnet 5.5 | 105.7 ±5.6 | medium | Sep 2026 | Nov 2026 | Official leaderboard |
| 5 | GPT-6.1 Terra | 103.7 ±4.8 | medium | Oct 2026 | Oct 2026 | Official leaderboard |
| 6 | | 103.3 ±12 | low | Aug 2026 | Nov 2026 | Official leaderboard |
| 7 | | 102.9 ±6.9 | medium | Nov 2026 | Nov 2026 | Official leaderboard |
| 8 | GPT-6.1 Sol | 101.6 ±13.4 | medium | Sep 2026 | Oct 2026 | Official leaderboard |
| 9 | Claude Haiku 5.5 | 98.7 ±11.7 | medium | Oct 2026 | Nov 2026 | Official leaderboard |
| 10 | GPT-6.2 Astra | 98.7 ±14 | medium | Nov 2026 | Nov 2026 | Official leaderboard |
| 11 | | 98.2 ±37.9 | none · Provisional | Mar 2025 | Nov 2026 | Official leaderboard |
| 12 | GPT-6 Luna | 95.2 ±4.4 | medium | Sep 2026 | Oct 2026 | Official leaderboard |
| 13 | GPT-6 Sol | 90.0 ±10.5 | medium | Sep 2026 | Oct 2026 | Official leaderboard |
| 14 | | 0.0 0 to 0.1 | none · Provisional | Feb 2026 | Nov 2026 | Official leaderboard |
Rankings by part
We read 3 parts from the benchmark’s own leaderboard: Daily opening, Fresh jar and Midgame. Each ranks models on one part of the test. Gemini 4 Neon leads 2 of them, and Gemma 3 4B leads the other one. Arrows show how far a model moved from its overall rank among the models we track.
Daily opening
14 models we track have a score here.
- 1
Gemini 4 Neon 117.1
- 2 GPT-6 Astra 109.3
- 3 GPT-6.1 Terra ↑2 108.8
- 4 GPT-6.1 Sol ↑4 104.5
- 5 GPT-6 Luna ↑7 101.2
- 6 Claude Haiku 5.5 ↑3 99.5
- 7
Qwen3.8-Flash-Next ↓1 96.8
- 8 GPT-6 Sol ↑5 95.9
- 9 GPT-6.2 Astra ↑1 95.1
- 10
Gemini 4 Hydrogen ↓3 94.9
Fresh jar
14 models we track have a score here.
- 1
Gemini 4 Neon 128.3
- 2 Claude Sonnet 5.5 ↑2 107.4
- 3
Qwen 3.8 27B 105.0
- 4
Qwen3.8-Flash-Next ↑2 102.1
- 5
Gemini 4 Hydrogen ↑2 101.6
- 6 GPT-6.1 Terra ↓1 101.5
- 7 GPT-6 Astra ↓5 98.7
- 8 Claude Haiku 5.5 ↑1 94.7
- 9 GPT-6.1 Sol ↓1 93.6
- 10 GPT-6 Luna ↑2 91.6
Midgame
14 models we track have a score here.
- 1
Gemma 3 4B ↑10 134.1
- 2 GPT-6 Astra 132.9
- 3
Qwen 3.8 27B 119.9
- 4 GPT-6.2 Astra ↑6 113.6
- 5 GPT-6.1 Sol ↑3 111.8
- 6 Claude Sonnet 5.5 ↓2 109.8
- 7
Gemini 4 Hydrogen 108.2
- 8
Qwen3.8-Flash-Next ↓2 107.7
- 9 GPT-6.1 Terra ↓4 104.6
- 10
Gemini 4 Neon ↓9 104.2
The best score, 114.5, costs $6.45 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: Official Crest Arcade Starjar leaderboard
| Model | Effort | Harness | Score | Cost per task | Output tokens |
|---|---|---|---|---|---|
| medium · Provisional | 114.5 | $6.45 | |||
| GPT-6 Astra | medium | 112.3 | $3.56 | ||
| none · Provisional | 108.1 | ||||
| Claude Sonnet 5.5 | medium | 105.7 | $0.68 | ||
| GPT-6.1 Terra | medium | 103.7 | $0.27 | ||
| low | 103.3 | ||||
| medium | 102.9 | $2.83 | |||
| GPT-6.1 Sol | medium | 101.6 | $0.89 | ||
| Claude Haiku 5.5 | medium | 98.7 | $0.09 | ||
| GPT-6.2 Astra | medium | 98.7 | $1.60 | ||
| none · Provisional | 98.2 | ||||
| GPT-6 Luna | medium | 95.2 | $0.04 | ||
| GPT-6 Sol | medium | 90.0 | $1.00 | ||
| none · Provisional | 0.0 |
The best score, 114.5, costs $6.45 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: Official Crest Arcade Starjar leaderboard
| Model | Effort | Harness | Score | Cost per task | Output tokens |
|---|---|---|---|---|---|
| medium · Provisional | 114.5 | $6.45 | |||
| GPT-6 Astra | medium | 112.3 | $3.56 | ||
| none · Provisional | 108.1 | ||||
| Claude Sonnet 5.5 | medium | 105.7 | $0.68 | ||
| GPT-6.1 Terra | medium | 103.7 | $0.27 | ||
| low | 103.3 | ||||
| medium | 102.9 | $2.83 | |||
| GPT-6.1 Sol | medium | 101.6 | $0.89 | ||
| Claude Haiku 5.5 | medium | 98.7 | $0.09 | ||
| GPT-6.2 Astra | medium | 98.7 | $1.60 | ||
| none · Provisional | 98.2 | ||||
| GPT-6 Luna | medium | 95.2 | $0.04 | ||
| GPT-6 Sol | medium | 90.0 | $1.00 | ||
| none · Provisional | 0.0 |
4 models have held number one on the official leaderboard since 2023
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official Crest Arcade Starjar leaderboard
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official Crest Arcade Starjar 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 Games and long tasks
- ARC-AGI-3 In use Interactive game environments where agents must explore, infer goals and rules, and act efficiently without instructions. GPT-6.2 Astra 100%
- Orak Fading LLM agents play 12 popular video games across most genres through an MCP interface. Each game scores in-game progress or outcomes, and the board ranks models by their average rank across the games.
Gemini 2.5 Pro 3.90
- MineDojo Fading Agents complete open-ended tasks in Minecraft, such as mining blocks and fighting mobs, from a large simulation suite. Programmatic checks score many tasks.
Seed2.0 Pro 49%
- TextQuests Only in system cards Agents play 25 classic interactive fiction games. The score is harmful actions per 100 steps, where lower is better. The board also reports game progress.
Llama 4 Scout 7.00
- GameWorld Only in system cards Multimodal game agents play 34 browser games across 170 tasks. Outcomes are checked from the game state. GPT-6.1 Terra 76.8%
- Swarm Bench Only in system cards Models orchestrate agent swarms on complex tasks that benefit from coordinated decomposition and parallel execution. Kimi K3 76.3%
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 official leaderboard (15 entries).
- Pass Models scored 14 models we track, plus 1 other entry on its leaderboard.
- Pass Newest score The newest is from November 2026, for Gemini 4 Neon.
- Pass Frontier models 5 of the top 10 have a score, led by GPT-6.2 Astra (#1).
- Note Headroom Scored as an index, with no fixed ceiling.
Leaderboard import 3 pass
- Pass Importer We read the board on glimmerindex.com: 14 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.