impulse

Nova 2

Transcription
API id nova-2
Input / 1M
-
not listed
Output / 1M
-
not listed
Context window
-
tokens
Max output
-
tokens per reply
Hosts
1
1 listings
Released
-
date not published

Where to run it

1 host serve Nova 2. The comparison is on blended price, 3 parts input to 1 part output.

HostIn / out per 1M
nova-2
-

Capabilities

Core

  • Streaming
  • Function calling
  • JSON mode
  • Structured outputs

Multimodal

  • Vision
  • Image input
  • Audio
  • Image generation
  • Video

Advanced

  • Reasoning
  • Tool calling
  • MCP
  • Prompt caching
  • Batch API
  • Embeddings
Takes in
Text
Produces
Text

Similar models

This model has no listed price, so these are current models with similar capabilities.

Smaller Amazon models

No similar Amazon model keeps its key capabilities.

Other frontier labs

Open-weight picks

Model line

Newer and older versions, at each lab's own price.

  1. UndatedNova 3LatestNot listed
  2. UndatedNova 2This pageNot listed

Call it

Using Deepgram's API.

Python
from litellm import completion

response = completion(
    model="deepgram/nova-2",
    messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)

Quick answers

Which APIs offer Nova 2?

Nova 2 runs on 1 API: Deepgram.

Is Nova 2 being retired?

No retirement date is scheduled for it on any API we track.

Is there a newer version of Nova 2?

Yes. Nova 3 is the newest model in this line.

What is Nova 2's API model id?

Use nova-2 with Deepgram.

Sources

Raw catalog entry for nova-2 on Deepgram
{
  "input_cost_per_second": 0.00007167,
  "litellm_provider": "deepgram",
  "metadata": {
    "calculation": "$0.0043/60 seconds = $0.00007167 per second",
    "original_pricing_per_minute": 0.0043
  },
  "mode": "audio_transcription",
  "output_cost_per_second": 0,
  "source": "https://deepgram.com/pricing",
  "supported_endpoints": [
    "/v1/audio/transcriptions"
  ]
}