GitHub[208K]Try OpenCode

Compare AI models: DeepSeek V4 Pro vs nemotron-3-ultra

  1. DeepSeek V4 ProDeepSeek
  2. nemotron-3-ultraNVIDIAHollow points use fallbacks · hover for details
ReasoningCodingCost efficiencyContext windowMultimodalTool use
Normalized model capability scores
ModelReasoningCodingCost efficiencyContext windowMultimodalTool use
DeepSeek V4 Pro100/100100/10033/10082/1000/10068/100
nemotron-3-ultra50/100 — No data; neutral placeholder50/100 — No data; neutral placeholder50/100 — No data; neutral placeholder50/100 — No data; neutral placeholder50/100 — No data; neutral placeholder50/100 — No data; neutral placeholder
Overview
Author
Context length
1M
Unknown
Reasoning
True
Unknown
Input modalities
Text
Unknown
Output modalities
Text
Unknown
Providers
Pricing
Input
$1.74/ 1M
Unknown
Output
$3.48/ 1M
Unknown
Cached input
$0.20/ 1M
Unknown
MomentumLast 2 mo
Unique users
603K
432K
Completed sessions
1,413,683
10,751,392
Token share
0.74%
2.83%
Tokens
7.4T-79%
28T+1,149%
AUG 20OCT 14
AUG 20OCT 14
RetentionWeek 1
Returning users
60.7%56K user-weeks
Pending

Related comparisons. Other model pairs to check.