Classification
Draft ICD-10 Code Suggestion
Given synthetic documented diagnoses and a supplied code subset, rank candidate codes and cite the exact chart language; always label output as coding-review support.
Healthcare & Life SciencesPer chart
Not benchmarked
No result on this tier
Token profile pending
— /100K
Run this tier to add it
—
Ministral 8B Instruct 2410
8B · Off-the-shelf
324 in · 180 out · 4.58s p50
$7.56 /100K
AWS Bedrock temporary proxy price
View token rates
AWS Bedrock · $0.1500 input · $0.1500 output /1M
62%
Mistral Large 2407
123B · Off-the-shelf
378 in · 217 out · 5.46s p50
$54.28 /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
83%
GPT-5.4
Hosted · Off-the-shelf
316 in · 226 out · 2.64s p50
$418 /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
99%
Best accuracy
All 12 model results and methodology
Generated 100/100 examples
| Model | Tier | Quality | Judged | Scenario cost |
|---|---|---|---|---|
| Nemotron Nano 9B v2 | small | 48% | 100/100 | $16.93 |
| Qwen3 235B A22B | mid | 76% | 100/100 | $53.92 |
| DeepSeek V3 | mid | 21% | 100/100 | $18.09 |
| Mistral Large 2407 | mid | 83% | 100/100 | $54.28 |
| Arcee Trinity Large Thinking | mid | 56% | 100/100 | $92.93 |
| GPT-5.4 | frontier | 99% | 100/100 | $418 |
| Claude Opus 4.7 | frontier | 91% | 100/100 | $1207 |
| Gemini 3.1 Pro Preview | frontier | 68% | 100/100 | $754 |
| Gemma 4 E4B IT | small | 34% | 100/100 | $1.38 |
| Granite 4.1 8B | small | 43% | 100/100 | $3.06 |
| Ministral 8B Instruct 2410 | small | 62% | 100/100 | $7.56 |
| Qwen3 4B Instruct 2507 | small | 53% | 100/100 | $16.33 |
LLM-judge pass rate on 100 synthetic examples. Generator: gpt-5.2. Judge: gpt-5.2. Evaluated 2026-09-09T07:10:02.552Z.
Directional: measured on a synthetic eval set generated by drydock. Cost/latency are not yet captured for taskrouter-run benchmarks.
drydock