Deepseek (latest) on inference.net
Deepseek (latest) through your own inference.net key. Price, speed and policy are this upstream's.
Performance and pricing
- Latency (TTFT)
- —
- Throughput
- —
- Uptime
- —
- Input
- $0 /M
- Output
- $0 /M
- Context
- 1.0M tokens
Billed by your own inference.net key. These rates show in your usage logs and are not charged to your AnyRouter credits.
Request example
curl https://anyrouter.dev/api/v1/chat/completions \
-H "Authorization: Bearer $ANYROUTER_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "deepseek/deepseek-latest",
"messages": [
{
"role": "user",
"content": "Say hi in 3 words."
}
],
"provider": {
"only": [
"inference-net-byok"
]
}
}'anyr claude --model deepseek/deepseek-latestUses your API key. Works with more tools
Performance over time
No history yet. Charts fill in as requests arrive.
Usage
Loading usage…
Usage over time
No usage yet.
Details
- Max output
- 131K tokens
- Region
- —
- Upstream model name
deepseek-v4-pro-0813- Billing
- Your own key (BYOK)
Features
No ZDRToolsReasoning
Data policy
No ZDR — data may be retained