2026-W34: Vibe coding
Week 34 of 2026 graded 12 coding agents on vibe coding. Gemini 3.8 Flash leads with B-. The biggest move since W33 2026 is Grok 4.7, C+ to D. Grades compare only within the same suite version, and plan usage is each agent's share of its own subscription. Every agent ran the same suite cut headless through its own tools.Measured
vibe-coding@v1 - the suite key these grades compare within
Published suite snapshot for 12 graded agents, current through Aug 18, 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 |
|---|---|---|---|---|---|---|
| Gemini 3.8 Flashantigravity-cli | B- | 79.4% | level 4 · partial level 5 | 4m 13s | 100% | 0.6%/1wGoogle AI Pro |
| Claude Fable 5.1claude-code | C+ | 69.0% | level 3 · partial level 5 | 5m 25s | 100% | 11.2%/5hClaude Max 5x |
| Claude Opus 5.5claude-code | C+ | 69.0% | level 3 · partial level 5 | 5m 14s | 100% | 3.2%/5hClaude Max 5x |
| GPT-5.6 Terracodex-cli | C- | 56.1% | level 2 · partial level 5 | 2m 46s | 100% | 0.5%/1wChatGPT Plus |
| GPT-6.1 Solcodex-cli | C- | 56.1% | level 2 · partial level 5 | 3m 43s | 100% | 0.8%/1wChatGPT Plus |
| Grok 4.7grok-cli | D | 45.8% | level 2 · partial level 5 | 8m 23s | 100% | 0.1%/1wSuperGrok Plus |
| Kimi K3 256Kkimi-cli | D- | 38.1% | level 2 · partial level 5 | 6m 37s | 100% | 10.2%/5hKimi Allegretto |
| GPT-6 Lunacodex-cli | D- | 36.8% | level 1 · partial level 5 | 2m 44s | 100% | 0.0%/1wChatGPT Plus |
| Composer 2.5cursor-cli | D- | 36.8% | level 1 · partial level 5 | 1m 49s | 100% | 0.1%/1mCursor Pro |
| Claude Sonnet 5.5claude-code | F | 34.2% | level 1 · partial level 5 | 5m 17s | 100% | 2.4%/5hClaude Max 5x |
| Gemini 3.1 Proantigravity-cli | F | 31.6% | none | 4m 31s | 100% | 0.9%/1wGoogle AI Pro |
| Claude Haiku 4.5claude-code | F | 23.2% | none | 4m 18s | 100% | 0%/5hClaude Max 5x |
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.