Head to head

Grok 4.5 vs Qwen3.8 2.4T-A95B

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.

Grok 4.5Qwen3.8 2.4T-A95B
Provider xAI Together AI
Input / MTok $2 $2.5
Output / MTok $6 $6.25
Cache read / MTok $0.3
Context window 500,000
Blended 1M in + 3M out $20.00 $21.25
Status ga ga

Grok 4.5 detail   Qwen3.8 2.4T-A95B detail