Published task benchmark

Dependency Upgrade Agent

Upgrade a dependency, migrate breaking APIs/config, and validate security and behavior.

Compare measured quality, workload cost, and latency across model tiers. Results are task-specific and directional—not a general model ranking.

Software engineering, data, and DevOpsAnalysisEvaluated 9/11/2026100 evaluation examplesFallback pricing checked September 11, 2026
Analysis

Dependency Upgrade Agent

Upgrade a dependency, migrate breaking APIs/config, and validate security and behavior.

Software engineering, data, and DevOpsPer dependency
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
597 in · 1,485 out · 24.64s p50
$16.93 /100K
Current price · TrustedRouter
View token rates · observed 9/11/2026
TrustedRouter · $0.0211 input · $0.1055 output /1M
71%
DeepSeek V3
Hosted · Off-the-shelf
567 in · 1,172 out · 16.24s p50
$129 /100K
Best current price · TrustedRouter
Compare 2 router rates · observed 9/11/2026
TrustedRouter · $0.3376 input · $0.9390 output /1M
OpenRouter · $0.2500 input · $1.00 output /1M
47%
GPT-5.4
Hosted · Off-the-shelf
540 in · 1,996 out · 18.89s p50
$3129 /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
28%
All 12 model results and methodology

Generated 100/100 examples

ModelTierQualityJudgedScenario cost
Nemotron Nano 9B v2small1%100/100$48.98
Qwen3 235B A22Bmid26%100/100$362
DeepSeek V3mid47%100/100$129
Mistral Large 2407mid34%100/100$232
Arcee Trinity Large Thinkingmid11%100/100$166
GPT-5.4frontier28%100/100$3129
Claude Opus 4.7frontier3%100/100$5464
Gemini 3.1 Pro Previewfrontier0%100/100$2512
Gemma 4 E4B ITsmall71%100/100$16.93
Granite 4.1 8Bsmall25%100/100$14.90
Ministral 8B Instruct 2410small17%100/100$25.18
Qwen3 4B Instruct 2507small3%100/100$130

LLM-judge pass rate on 100 synthetic examples. Generator: gpt-5.2. Judge: gpt-5.2. Evaluated 2026-09-11T09:58:02.615Z.

Directional: measured on a synthetic eval set generated by drydock. Cost/latency are not yet captured for taskrouter-run benchmarks.

drydock

Have a neighboring workload?

Describe it in Task Explorer. If it is not measured yet, you can request a benchmark and follow future results.

Explore another task