DeepSeek V3 0324
deepinfra/deepseek-ai/DeepSeek-V3-0324 · family deepseek · in: text → out: text · price provider-declared (via models.dev). Capabilities: toolstext.
$0.41Blended $/1M (3:1)
$0.24Input $/1M
$0.9Output $/1M
$1.141M in + 1M out
$0.135 −44%Cache read $/1M
#1896 of 5,861Cheapest rank
- Provider
- Deep Infra (
deepinfra) - Kind
- text (outputs text)
- Context
- 163,840 tokens
- Max output
- 163,840 tokens
- Capabilities
- toolstext
Same model, other providers
“DeepSeek V3 0324” is offered by 9 providers — cheapest-effective first. submodel is the cheapest at $0.35 blended.
| Provider | Input | Output | Blended | Context |
|---|---|---|---|---|
| submodel | $0.2 | $0.8 | $0.35 | 75k |
| Deep Infra (this) | $0.24 | $0.9 | $0.41 | 164k |
| Meganova | $0.25 | $0.88 | $0.41 | 164k |
| Vercel AI Gateway | $0.27 | $1.12 | $0.48 | 164k |
| NovitaAI | $0.27 | $1.12 | $0.48 | 164k |
| OpenRouter | $0.27 | $1.12 | $0.48 | 164k |
| Hugging Face | $0.27 | $1.12 | $0.48 | 164k |
| Jiekou.AI | $0.28 | $1.14 | $0.49 | 164k |
| Crusoe | $0.5 | $1.5 | $0.75 | 164k |
Call it
jishie indexes & prices models; it does not proxy inference.
Most providers are OpenAI-compatible — point base_url at Deep Infra 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-V3-0324","messages":[{"role":"user","content":"hello"}]}'
Route programmatically with MCP find_model / get_model, or fetch this record free (no wallet):