Deepgram Nova-3 vs Gradium
Accuracy and speed
Left gets fewer words wrong. Lower has the final text ready sooner.
- Nova-33.46% · 103ms
- Gradium6.48% · 212ms
Deepgram Nova-3 is ahead of Gradium on both datasets together, the Pipecat Dataset, time to first text and final-text delay. The Ocular Dataset is too close to call.
Last updated Methodology by Dileep Chagam
Who is better at what
- Overall WER3.78%5.78%
- Pipecat Dataset WER3.46%6.48%
- Ocular Dataset WER4.39%4.45%
- Time to first text1.06s1.71s
- Final-text delay103ms212ms
Lower is better on every measure. Too close to call means the likely ranges overlap. For costs, see our breakdown of Deepgram pricing.
Nova-3 vs Gradium FAQ
Which is more accurate, Deepgram Nova-3 or Gradium?
Deepgram Nova-3 is more accurate on both datasets. On the 1,000 public Pipecat clips, Deepgram Nova-3 scores 3.78% and Gradium 5.78%, so Deepgram Nova-3 makes fewer errors by 1.99 points. On the eight licensed Ocular recordings, Deepgram Nova-3 scores 3.46% and Gradium 6.48%, so Deepgram Nova-3 makes fewer errors by 3.01 points. Word error rate counts incorrect, missing, and extra words, so lower is better.
Deepgram Nova-3 is more accurate on both datasets. On the 1,000 public Pipecat clips, Deepgram Nova-3 scores 3.78% and Gradium 5.78%, so Deepgram Nova-3 makes fewer errors by 1.99 points. On the eight licensed Ocular recordings, Deepgram Nova-3 scores 3.46% and Gradium 6.48%, so Deepgram Nova-3 makes fewer errors by 3.01 points. Word error rate counts incorrect, missing, and extra words, so lower is better.
Yes. Deepgram Nova-3 finishes the transcript 641ms sooner. Deepgram Nova-3 has the final transcript 1.06s after the speaker stops and Gradium 1.71s, at the median. Both are measured on the same 206 streamed turns.
Deepgram Nova-3 is the stronger pick for voice agents: it makes fewer errors on real conversations and finishes the transcript 641ms sooner. A voice agent cannot reply until the final transcript arrives, and the Ocular recordings are real conversations.
Yes, both return partial text before the speaker stops. Deepgram Nova-3 shows first text after 4.39% and Gradium after 4.45%, at the median.
Test your voice agent the same way.
Cekura runs these simulated calls on your agent and scores every one.