Nvidia Parakeet TDT 0.6B V3
Open weightsNvidia Parakeet TDT 0.6B V3 is a transcription model from Nvidia with a 0K-token context window. Provider list price is $0.0015 per minute; on Allocate you pay $0.0016 with the 7% transaction fee. The weights are open, so you can fine-tune it and own the result.
Pricing
Token usage bills at the provider list price plus the 7% transaction fee. Prices checked 2026-07-08.
Price against its peers
Provider list prices per minute, Nvidia Parakeet TDT 0.6B V3 against its nearest transcription peers by price.
What a real workload costs
Transcribing 1,000 minutes of audio costs 1,000 × $0.0015 = $1.50 at list, or $1.60 billed on Allocate with the 7% transaction fee included.
Nvidia Parakeet TDT 0.6B V3 is an open-weights model; the catalog does not list its license, so check the lab’s model card before commercial fine-tuning.
Example usage
Point a route at nvidia/parakeet-tdt-0.6b-v3 and the endpoint never changes; swap the model behind it whenever you want.
curl https://api.allocate.network/v1/chat/completions \ -H "Authorization: Bearer $ALLOCATE_KEY" \ -d '{ "model": "nvidia/parakeet-tdt-0.6b-v3", "messages": [{"role": "user", "content": "Summarise the attached contract."}] }'
Common questions
How much does Nvidia Parakeet TDT 0.6B V3 cost?
Provider list price is $0.0015 per minute. On Allocate you pay list plus the 7% transaction fee: $0.0016.
What context window does Nvidia Parakeet TDT 0.6B V3 have?
448 tokens (0K). At roughly 0.75 words per token, that is about 0k words of English text per request.
Can I fine-tune Nvidia Parakeet TDT 0.6B V3?
Yes. Nvidia Parakeet TDT 0.6B V3 is an open-weights model; check the lab’s model card for the exact license terms. Read the license terms before fine-tuning for commercial use. On Allocate the trained weights stay inside your boundary and belong to you.
How do I call Nvidia Parakeet TDT 0.6B V3 on Allocate?
Send nvidia/parakeet-tdt-0.6b-v3 in the model field of the OpenAI-compatible endpoint at api.allocate.network/v1, or point a route name (like prod/support-agent) at it so you can swap the model later without a deploy.