CoreWeave models

0 models in this catalog are created by CoreWeave; 1 additional models are available through its hosting offerings. The newest dated release listed is DeepSeek V4.1 Flash (Sep 10, 2026).

Updated: Nov 5, 2026Data as of: Nov 2, 2026

Visit CoreWeave

Compare hosting providers

DeepSeek V4.1 Flash(max)

Token cost

USD per 1M tokens, 50% input + 50% output. Lower is better.

$0.28
DeepInfra
$0.36
Bitdeer AI
$0.38
SiliconFlow
$0.43
CoreWeave
$0.45
GMI
$0.72
Makora
$0.75
Crusoe
$0.75
Baseten
$0.75
Databricks
$0.75
DigitalOcean
$0.75
Modal
$0.75
Nebius
$0.75
Novita
$0.75
Parasail
$0.75
Together AI
$0.75
Wafer
$1.4
Scaleway

Output speed

Tokens per second. Higher is better.

384.07 tok/s
Databricks
373.03 tok/s
Together AI
320.5 tok/s
Wafer
297.86 tok/s
CoreWeave
289.73 tok/s
Bitdeer AI
257.08 tok/s
Nebius
244.98 tok/s
Baseten
239.17 tok/s
Parasail
205.63 tok/s
Crusoe
201.35 tok/s
Scaleway
190.72 tok/s
Makora
164.07 tok/s
DeepInfra
157.79 tok/s
Modal
154.52 tok/s
SiliconFlow
140.7 tok/s
GMI
136.76 tok/s
Novita
81.9 tok/s
DigitalOcean

Time to first token

Seconds until the first token. Lower is better.

0.31 s
Together AI
0.56 s
Modal
0.56 s
Wafer
0.63 s
Baseten
0.63 s
Parasail
0.66 s
Makora
0.7 s
DeepInfra
0.77 s
Crusoe
0.82 s
DigitalOcean
0.91 s
Databricks
1 s
CoreWeave
1.23 s
Novita
1.33 s
Nebius
1.41 s
Scaleway
1.51 s
Bitdeer AI
1.6 s
SiliconFlow
1.81 s
GMI

All models

1 models

All CoreWeave models in the dataset, newest reported release dates first.
Model	Date released	Model details
DeepSeek V4.1 Flash	Sep 10, 2026	View model