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.

Claude Haiku 4.5Qwen3 VL 8B Instruct
Claude Haiku 4.5

Anthropic

Qwen3 VL 8B Instruct

Alibaba (Qwen)

Pricing
Input $/1M$1.00$0.117
Output $/1M$5.00$0.455
Blended $/1M (3:1)$2.00$0.202
Cache read $/1M$0.10
Cache write $/1M$1.25
Quality indices
Intelligence Index29.68.4
Coding Index43.9
Math Index27.3
Agentic Index16.4
Speed & latency
Output speed (tok/s)141
Time to first token0.90s
Time to first answer token0.90s
Benchmarks (Artificial Analysis)
MMLU-Pro68.6%
GPQA Diamond42.7%
Humanity's Last Exam2.9%
LiveCodeBench33.2%
SciCode17.4%
AIME 202527.3%
IFBench32.3%
AA-LCR (Long Context Reasoning)15.3%
Terminal-Bench Hard2.3%
τ²-Bench (Telecom)29.2%
Design Arena (Elo)
Website1,149
UI components1,142
Data viz1,160
SVG1,072
3D1,131
Game dev1,155
Code categories1,148
ASCII art1,182
Specs
Context window200K262K
Max output tokens64K33K
Input modalitiestext, image, fileimage, text
TokenizerClaudeQwen3