Head to head

Qwen3.8 2.4T-A95B vs Grok 4.5

On a blended workload of one million input tokens and three million output tokens, Grok 4.5 is 1.1× cheaper than Qwen3.8 2.4T-A95B.

Qwen3.8 2.4T-A95BGrok 4.5
Provider Together AI xAI
Input / MTok $2.5 $2
Output / MTok $6.25 $6
Cache read / MTok $0.3
Context window 500,000
Blended 1M in + 3M out $21.25 $20.00
Status ga ga

Qwen3.8 2.4T-A95B detail   Grok 4.5 detail