DeepSeek-V3.2

friendli/deepseek-ai/DeepSeek-V3.2 · family deepseek · price provider-declared (via models.dev), USD per 1M tokens. Capabilities: toolsreasoningtext.

$0.75Blended $/1M (3:1)

$0.5Input $/1M

$1.5Output $/1M

$2.001M in + 1M out

$0.25 −50%Cache read $/1M

#2789 of 5,000Cheapest rank

Provider: Friendli (friendli)
Context: 163,840 tokens
Max output: 163,840 tokens
Capabilities: toolsreasoningtext

Same model, other providers

“DeepSeek-V3.2” is offered by 9 providers — cheapest-effective first. Deep Infra is the cheapest at $0.29 blended.

Provider	Input	Output	Blended	Context
Deep Infra	$0.26	$0.38	$0.29	164k
Poe	$0.27	$0.4	$0.30	128k
Hugging Face	$0.28	$0.4	$0.31	164k
Vivgrid	$0.28	$0.42	$0.32	128k
302.AI	$0.29	$0.43	$0.32	128k
Nebius Token Factory	$0.3	$0.45	$0.34	163k
Friendli (this)	$0.5	$1.5	$0.75	164k
Azure Cognitive Services	$0.58	$1.68	$0.85	128k
Azure	$0.58	$1.68	$0.85	128k

Call it

jishie indexes & prices models; it does not proxy inference. Most providers are OpenAI-compatible — point base_url at Friendli and pass this model id:

curl $BASE_URL/v1/chat/completions \
  -H "Authorization: Bearer $API_KEY" -H "Content-Type: application/json" \
  -d '{"model":"deepseek-ai/DeepSeek-V3.2","messages":[{"role":"user","content":"hello"}]}'

Route programmatically with MCP find_model / get_model, or fetch this record free (no wallet):