Zero-Downtime Schema Migration Planning
Given a current schema, target schema, traffic pattern, and deployment constraints, order expand-migrate-contract steps and flag unsafe operations.
View token rates · observed 9/11/2026
Compare 2 router rates · observed 9/11/2026
Compare 2 router rates · observed 9/11/2026
All 12 model results and methodology
Generated 100/100 examples
| Model | Tier | Quality | Judged | Scenario cost |
|---|---|---|---|---|
| Nemotron Nano 9B v2 | small | 0% | 100/100 | $48.25 |
| Qwen3 235B A22B | mid | 10% | 100/100 | $281 |
| DeepSeek V3 | mid | 13% | 100/100 | $120 |
| Mistral Large 2407 | mid | 6% | 100/100 | $175 |
| Arcee Trinity Large Thinking | mid | 1% | 100/100 | $159 |
| GPT-5.4 | frontier | 91% | 100/100 | $2511 |
| Claude Opus 4.7 | frontier | 10% | 100/100 | $5339 |
| Gemini 3.1 Pro Preview | frontier | 3% | 100/100 | $2459 |
| Gemma 4 E4B IT | small | 17% | 100/100 | $14.16 |
| Granite 4.1 8B | small | 1% | 100/100 | $14.45 |
| Ministral 8B Instruct 2410 | small | 0% | 100/100 | $19.45 |
| Qwen3 4B Instruct 2507 | small | 0% | 100/100 | $114 |
LLM-judge pass rate on 100 synthetic examples. Generator: gpt-5.2. Judge: gpt-5.2. Evaluated 2026-09-09T02:35:02.507Z.
Directional: measured on a synthetic eval set generated by drydock. Cost/latency are not yet captured for taskrouter-run benchmarks.
drydock