2026-W36: Vibe coding
vibe-coding@v1 - the suite key these grades compare within
Published suite snapshot for 12 graded agents, current through Sep 1, 2026.
Sorted by capability. Plan usage is each agent's share of its own named subscription and is not a cross-agent ranking. See the Vibe coding methodology or each agent's scorecard for full context.
| Agent | Grade | Capability | Clears | Task time | Reliability | Plan usage |
|---|---|---|---|---|---|---|
| claude-fable-5claude-code | C | 74.2% | level 3 · partial level 5 | 5m 25s | 100% | 2.8%/5hClaude Max 20x |
| claude-opus-5claude-code | C | 69.0% | level 3 · partial level 5 | 5m 16s | 100% | 0.8%/5hClaude Max 20x |
| gemini-3.7-flashantigravity-cli | C | 61.3% | level 2 · partial level 5 | 3m 37s | 100% | 0.6%/1wGoogle AI Pro |
| gpt-5.6-solcodex-cli | C | 56.1% | level 2 · partial level 5 | 6m 12s | 100% | 0.7%/1wChatGPT Plus |
| kimi-k3-256kkimi-cli | D | 54.8% | level 1 · partial level 5 | 6m 25s | 100% | 11%/5hKimi Allegretto |
| gpt-5.6-lunacodex-cli | D | 51.0% | level 2 · partial level 5 | 2m 49s | 100% | 0.1%/1wChatGPT Plus |
| grok-4.6grok-cli | D | 45.8% | level 2 · partial level 5 | 8m 41s | 100% | 0.2%/1wSuperGrok |
| gemini-3.1-proantigravity-cli | D | 40.0% | none | 4m 8s | 100% | 0.7%/1wGoogle AI Pro |
| claude-haiku-4-5claude-code | D | 40.0% | none | 4m 33s | 100% | 0.2%/5hClaude Max 20x |
| composer-2.5cursor-cli | F | 34.8% | none | 2m 14s | 100% | 0.1%/1mCursor Pro |
| claude-sonnet-5claude-code | F | 34.2% | level 1 · partial level 5 | 4m 50s | 100% | 0.2%/5hClaude Max 20x |
| gpt-5.6-terracodex-cli | F | 29.0% | level 1 · partial level 5 | 2m 44s | 100% | 0.5%/1wChatGPT Plus |
This page publishes aggregate outcomes only. Task prompts and private checker logic never leave the benchmark. Follow this suite over time on Grades over time.