Generation
Unit-Test Case Generation
Given a function contract and implementation, generate executable unit-test cases covering happy paths, boundaries, invalid inputs, and observed failure branches.
Engineering & ProductPer function
Not benchmarked
No result on this tier
Token profile pending
— /100K
Run this tier to add it
—
Gemma 4 E4B IT
E4B · Off-the-shelf
351 in · 943 out · 15.17s p50
$10.69 /100K
Current price · TrustedRouter
View token rates · observed 9/11/2026
TrustedRouter · $0.0211 input · $0.1055 output /1M
69%
Cheapest
DeepSeek V3
Hosted · Off-the-shelf
312 in · 869 out · 7.85s p50
$92.13 /100K
Best current price · TrustedRouter
Compare 2 router rates · observed 9/11/2026
TrustedRouter · $0.3376 input · $0.9390 output /1M
OpenRouter · $0.2900 input · $1.14 output /1M
58%
GPT-5.4
Hosted · Off-the-shelf
312 in · 667 out · 4.78s p50
$1079 /100K
Best current price · OpenRouter
Compare 2 router rates · observed 9/11/2026
OpenRouter · $2.50 input · $15.00 output /1M
TrustedRouter · $2.64 input · $15.82 output /1M
93%
Best accuracy
All 12 model results and methodology
Generated 100/100 examples
| Model | Tier | Quality | Judged | Scenario cost |
|---|---|---|---|---|
| Nemotron Nano 9B v2 | small | 1% | 100/100 | $47.28 |
| Qwen3 235B A22B | mid | 55% | 100/100 | $162 |
| DeepSeek V3 | mid | 58% | 100/100 | $92.13 |
| Mistral Large 2407 | mid | 58% | 100/100 | $147 |
| Arcee Trinity Large Thinking | mid | 24% | 100/100 | $141 |
| GPT-5.4 | frontier | 93% | 100/100 | $1079 |
| Claude Opus 4.7 | frontier | 43% | 100/100 | $4849 |
| Gemini 3.1 Pro Preview | frontier | 45% | 100/100 | $2232 |
| Gemma 4 E4B IT | small | 69% | 100/100 | $10.69 |
| Granite 4.1 8B | small | 31% | 100/100 | $10.95 |
| Ministral 8B Instruct 2410 | small | 21% | 100/100 | $16.80 |
| Qwen3 4B Instruct 2507 | small | 23% | 100/100 | $61.91 |
LLM-judge pass rate on 100 synthetic examples. Generator: gpt-5.2. Judge: gpt-5.2. Evaluated 2026-09-10T08:57:02.527Z.
Directional: measured on a synthetic eval set generated by drydock. Cost/latency are not yet captured for taskrouter-run benchmarks.
drydock