Deepgram Nova-3 vs Gradium
Deepgram Nova-3 is ahead of Gradium on all four metrics.
Quality and speed frontier
Every model's error rate on the public clips, against how long its transcript takes to settle.
The four metrics, side by side
Word error rate on 1,000 clips from Pipecat's STT benchmark dataset.
- Deepgram Nova-3
- 3.46%10th
- Gradium
- 6.48%15th
Deepgram Nova-3, 3.01 points lower
Word error rate on eight licensed recordings provided by Ocular.
- Deepgram Nova-3
- 4.39%10th
- Gradium
- 4.45%11th
Deepgram Nova-3, 0.06 points lower
Median delay before the first words arrive, from the first audio packet.
- Deepgram Nova-3
- 1.06s6th
- Gradium
- 1.71s10th
Deepgram Nova-3, 641ms sooner
Median delay after the speaker stops before the transcript is final.
- Deepgram Nova-3
- 103ms2nd
- Gradium
- 212ms9th
Deepgram Nova-3, 109ms sooner
Lower is better on every metric. The small figure is each model's place in the whole field.
First text and final text, at three percentiles
Both measured on the same 206 streamed Ocular Dataset turns. P90 and P95 say what a caller meets on a bad turn, which a median hides.
| Measure | Model | P50 | P90 | P95 |
|---|---|---|---|---|
| Time to first text | Deepgram Nova-3 | 1.06s | 1.10s | 2.07s |
| Gradium | 1.71s | 4.05s | 7.32s | |
| Final-text delay | Deepgram Nova-3 | 103ms | 143ms | 155ms |
| Gradium | 212ms | 266ms | 280ms |
Both against every other model
Open on the metric where the two sit furthest apart, with the other three a tab away.
Final-text delay after the speaker stops. Lower is better.
| # | Model | Final text | turns | |
|---|---|---|---|---|
| 1 | Inworld STT-1 | 41ms | 206 | |
| 2 | Deepgram Nova-3 | 103ms | 202 | |
| 3 | Deepgram Flux English | 108ms | 199 | |
| 4 | Deepgram Flux Multilingual | 114ms | 175 | |
| 5 | Cartesia Ink 2 | 123ms | 198 | |
| 6 | Speechmatics Linden | 147ms | 194 | |
| 7 | AssemblyAI Universal 3.5 Pro | 180ms | 205 | |
| 8 | Smallest Pulse | 201ms | 206 | |
| 9 | Gradium | 212ms | 198 | |
| 10 | Reson8 | 235ms | 201 | |
| 11 | Google Chirp 3 | 475ms | 206 | |
| 12 | GPT Realtime Whisper | 543ms | 206 | |
| 13 | Google Chirp 2 | 568ms | 205 | |
| 14 | GPT-4o Mini Transcribe | 630ms | 206 | |
| 15 | GPT-4o Transcribe | 652ms | 206 |
Median across the same 206 Ocular Dataset turns. Providers finalize differently, so compare models with similar finalization contracts before treating small gaps as meaningful.
Both on the other two views
Pipecat vs Ocular accuracy
Every model's error rate on the public clips, against the same model on real conversations.
Transcription response time
How soon each model returns its first words, against how soon its transcript is final.
What the numbers rest on
| Model | Pipecat clips | Ocular recordings | Turns timed | Failed |
|---|---|---|---|---|
| Deepgram Nova-3 | 1,000 | 8 | 202 | 0 |
| Gradium | 1,000 | 8 | 198 | 0 |
A clip counts only where the model returned usable text, so counts differ between models. Word error rate is measured the same way throughout: incorrect, missing, and extra words against the reference transcript.
Frequently asked questions
Is Deepgram Nova-3 or Gradium more accurate?
On the 1,000 public Pipecat clips, Deepgram Nova-3 scores 3.46% and Gradium 6.48%, so Deepgram Nova-3 makes fewer errors by 3.01 points. On the eight licensed Ocular recordings, Deepgram Nova-3 scores 4.39% and Gradium 4.45%, so Deepgram Nova-3 makes fewer errors by 0.06 points. Word error rate counts incorrect, missing, and extra words against the reference transcript, so lower is better.
Which is faster, Deepgram Nova-3 or Gradium?
Deepgram Nova-3 returns first text 641ms sooner, at 1.06s against 1.71s at the median. Deepgram Nova-3 has the final transcript 109ms sooner, 103ms after the speaker stops against 212ms. Both are measured on the same 206 streamed turns.
How were Deepgram Nova-3 and Gradium compared?
Both ran the same two datasets under the same conditions as every other model in the benchmark: 1,000 public clips from Pipecat's STT benchmark dataset and eight licensed conversation recordings from Ocular. Deepgram Nova-3 returned usable text for 1,000 of the public clips and Gradium for 1,000. Latency uses each provider's own supported finalization contract.