AiCostCompare

Compare models

Pick up to 4 models for a side-by-side spec sheet. Hover any dotted metric name for what the test measures. Best value in each row is highlighted.

GPT-5.6 SolQwen3.6 35B A3B
GPT-5.6 Sol

OpenAI

Qwen3.6 35B A3B

Alibaba (Qwen)

Pricing
Input $/1M$2.00$0.10
Output $/1M$10.00$0.90
Blended $/1M (3:1)$4.00$0.30
Cache read $/1M$0.20$0.05
Cache write $/1M$2.50
Quality indices
Intelligence Index4718.2
Coding Index77.441.9
Math Index
Agentic Index50.213.1
Speed & latency
Output speed (tok/s)65114
Time to first token64.46s1.08s
Time to first answer token64.46s48.57s
Benchmarks (Artificial Analysis)
GPQA Diamond94.1%84.1%
Humanity's Last Exam49.5%22.2%
SciCode57.1%36.6%
IFBench72.7%64.4%
AA-LCR (Long Context Reasoning)84.0%71.7%
Terminal-Bench Hard65.9%34.8%
Terminal-Bench 2.188.0%44.9%
τ²-Bench (Telecom)85.1%95.3%
τ³-Bench Banking44.3%9.3%
Specs
Context window1.05M262K
Max output tokens128K236K
Input modalitiesfile, image, texttext, image, video
TokenizerGPTQwen