INDEX
Inference Exchange

The Compute
Exchange

LIVE

Buy inference from the cheapest provider.
Sell your compute to the highest bidder.
Prices move in real-time.

MODELS
CLUSTERS
Sell Compute →Get Compute
Model Architecture
Output Price (1M Tok)
Global Latency (TTFT)
Context Window
Providers
Action
Llama-3-70b-Instruct
MetaFP16
$0.79/ 1M
▼ 2.4%
240ms
8,192
Tokens
12
Active Nodes
Connect
Mixtral-8x22B
Mistral AIMoE
$0.65/ 1M
— 0.0%
310ms
32,768
Tokens
08
Active Nodes
Connect
Qwen1.5-110B-Chat
AlibabaInt8
$1.12/ 1M
▲ 1.1%
450ms
32,768
Tokens
04
Active Nodes
Connect
Claude-3-Haiku
AnthropicProxy
$0.25/ 1M
▼ 5.2%
180ms
200,000
Tokens
42
Active Nodes
Connect
DeepSeek-V3
DeepSeekFP16
$0.14/ 1M
▼ 8.1%
195ms
128,000
Tokens
18
Active Nodes
Connect
Gemma-2-27B-IT
GoogleBF16
$0.35/ 1M
▲ 3.2%
280ms
8,192
Tokens
15
Active Nodes
Connect