2026-W39: Real-world issues
Week 39 of 2026 graded 13 coding agents on real-world issues. GPT-5.6 Sol leads with C-. The biggest move since W38 2026 is Gemini 3.8 Flash, D+ to F. 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
real-world@v3 - the suite key these grades compare within
Published suite snapshot for 13 graded agents, current through Sep 21, 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 Real-world issues methodology or each agent's scorecard for full context.
| Agent | Grade | Capability | Clears | Task time | Reliability | Plan usage |
|---|---|---|---|---|---|---|
| GPT-5.6 Solcodex-cli | C- | 60.0% | tier 4 · partial tier 5 | 3m 13s | 100% | 2%/1wChatGPT Pro 5x |
| Claude Opus 5claude-code | C- | 56.7% | tier 2 · partial tier 5 | 1m 56s | 100% | 21%/5hClaude Max 5x |
| Kimi K3 256Kkimi-cli | C- | 56.7% | tier 2 · partial tier 5 | 4m 5s | 100% | 115%/5hKimi Allegretto |
| Gemini 3.1 Proantigravity-cli | D+ | 53.3% | tier 3 · partial tier 5 | 2m 55s | 100% | 9.8%/1wGoogle AI Pro |
| Claude Fable 5.1claude-code | D+ | 53.3% | tier 3 · partial tier 5 | 2m 20s | 100% | 100%/5hClaude Max 5x |
| Claude Haiku 4.5claude-code | D+ | 50.0% | tier 2 · partial tier 5 | 3m 39s | 100% | 13%/5hClaude Max 5x |
| Claude Sonnet 5claude-code | D+ | 50.0% | tier 2 · partial tier 5 | 2m 55s | 100% | 16%/5hClaude Max 5x |
| GPT-5.6 Terracodex-cli | D+ | 50.0% | tier 2 · partial tier 5 | 3m 29s | 100% | 2%/1wChatGPT Pro 5x |
| Composer 2.5cursor-cli | D+ | 50.0% | tier 2 · partial tier 5 | 1m 21s | 100% | 1.7%/1mCursor Pro |
| Grok 4.6grok-cli | D | 46.7% | tier 4 · partial tier 5 | 7m 10s | 100% | 3%/1wSuperGrok Plus |
| GPT-5.6 Lunacodex-cli | D | 45.0% | tier 1 · partial tier 5 | 3m 7s | 100% | 0%/1wChatGPT Pro 5x |
| GPT-6 Astracodex-cli | D | 43.3% | tier 2 · partial tier 5 | 2m 15s | 100% | 5%/1wChatGPT Pro 5x |
| Gemini 3.8 Flashantigravity-cli | F | 28.3% | none | 3m 11s | 100% | 18.0%/1wGoogle AI Pro |
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.