Nemotron 3 Nano 30B A3B
thinkingmachines/nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16 · family nemotron · in: text → out: text · price provider-declared (via models.dev). Capabilities: tools reasoning text.
$0.27
Blended $/1M (3:1)
$0.195
Input $/1M
$0.495
Output $/1M
$0.69
1M in + 1M out
$0.039 −80%
Cache read $/1M
#2396 of 8,682
Cheapest rank
- Provider
- Thinking Machines (
thinkingmachines) - Kind
- text (outputs text)
- Context
- 65,536 tokens
- Max output
- 65,536 tokens
- Capabilities
- tools reasoning text
Same model, other providers
“Nemotron 3 Nano 30B A3B” is offered by 10 providers — cheapest-effective first. Kenari is the cheapest at $0.00 blended.
| Provider | Input | Output | Blended | Context |
|---|---|---|---|---|
| Kenari | free | free | $0.00 | 262k |
| Merge Gateway | $0.05 | $0.2 | $0.09 | 256k |
| Vercel AI Gateway | $0.05 | $0.2 | $0.09 | 262k |
| Deep Infra | $0.05 | $0.2 | $0.09 | 262k |
| Kilo Gateway | $0.05 | $0.2 | $0.09 | 262k |
| Crusoe | $0.05 | $0.2 | $0.09 | 262k |
| Pioneer | $0.05 | $0.2 | $0.09 | 262k |
| Novita AI | $0.05 | $0.2 | $0.09 | 262k |
| OpenRouter | $0.06 | $0.24 | $0.10 | 262k |
| Thinking Machines (this) | $0.195 | $0.495 | $0.27 | 66k |
Call it
jishie indexes & prices models; it does not proxy inference.
Most providers are OpenAI-compatible — point base_url at Thinking Machines and pass this model id:
curl $BASE_URL/v1/chat/completions \
-H "Authorization: Bearer $API_KEY" -H "Content-Type: application/json" \
-d '{"model":"nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16","messages":[{"role":"user","content":"hello"}]}'
Route programmatically with MCP find_model / get_model, or fetch this record free (no wallet):