Published task benchmark

Loss-Review

Summarizes why a customer left based on historical communication.

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

Customer SuccessSummarizationEvaluated 9/10/2026100 evaluation examplesFallback pricing checked September 11, 2026
Summarization

Loss-Review

Summarizes why a customer left based on historical communication.

Customer SuccessPer account
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
697 in · 230 out · 3.61s p50
$3.90 /100K
Current price · TrustedRouter
View token rates · observed 9/11/2026
TrustedRouter · $0.0211 input · $0.1055 output /1M
94%
Cheapest
Mistral Large 2407
123B · Off-the-shelf
741 in · 380 out · 10.46s p50
$99.22 /100K
Best current price · TrustedRouter
Compare 2 router rates · observed 9/11/2026
TrustedRouter · $0.5275 input · $1.58 output /1M
OpenRouter · $2.00 input · $6.00 output /1M
92%
GPT-5.4
Hosted · Off-the-shelf
618 in · 389 out · 4.28s p50
$738 /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
100%
Best accuracy
All 12 model results and methodology

Generated 100/100 examples

ModelTierQualityJudgedScenario cost
Nemotron Nano 9B v2small33%100/100$19.60
Qwen3 235B A22Bmid69%100/100$104
DeepSeek V3mid83%100/100$41.15
Mistral Large 2407mid92%100/100$99.22
Arcee Trinity Large Thinkingmid66%100/100$97.06
GPT-5.4frontier100%100/100$738
Claude Opus 4.7frontier74%100/100$2499
Gemini 3.1 Pro Previewfrontier95%100/100$1878
Gemma 4 E4B ITsmall94%100/100$3.90
Granite 4.1 8Bsmall16%100/100$7.03
Ministral 8B Instruct 2410small55%100/100$15.49
Qwen3 4B Instruct 2507small40%100/100$34.61

LLM-judge pass rate on 100 synthetic examples. Generator: gpt-5.2. Judge: gpt-5.2. Evaluated 2026-09-10T14:32:28.276Z.

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