Compare / Nova Pro 1.0 vs DeepSeek V4 Pro 0423

Nova Pro 1.0 vs DeepSeek V4 Pro 0423

Which is cheaper, which is more capable, and which does better at each kind of work, from 12 results on 3 sources that measured both models.

Nova Pro 1.0
Amazon · profile
DeepSeek V4 Pro 0423
DeepSeek · profile
Results compared
12
Shared sources
3

Nova Pro 1.0 or DeepSeek V4 Pro 0423? The short answer

  • Nova Pro 1.0 is cheaper: $1.40 against $1.98 per million tokens, 29% less.
  • DeepSeek V4 Pro 0423 scores higher on the Artificial Analysis Intelligence Index: 36.0 against 7.0.
  • On our business benchmarks, Nova Pro 1.0 ranks higher for sticking to the facts.
  • On our business benchmarks, DeepSeek V4 Pro 0423 ranks higher for agents and tool use, reasoning and knowledge and writing.

At a glance

Nova Pro 1.0DeepSeek V4 Pro 0423
Price per million tokens$0.80 in · $3.20 out$1.32 in · $3.96 out
Blended (3 in : 1 out)$1.40$1.98
Cheapest host, blended$1.40 (Amazon Bedrock)$0.26 (StreamLake, FP8)
Intelligence Index7.036.0
Spring Prompt overall–31
Output speed, tokens a second–115
Context window300,000 tokens1,048,576 tokens
WeightsClosedOpen
Developer based inthe USChina

Prices are the developer's list price where we have read it, otherwise the typical host on OpenRouter; see every model's price. Intelligence and speed from Artificial Analysis.

Which is better for what

Each model's rank among models available today, for each kind of work, from our own benchmarks and licensed sources. The better rank is in green.

Use caseNova Pro 1.0DeepSeek V4 Pro 0423
Decks from an analysisTurning a finished analysis into a deck you could present as it is not measured 17 of 18
Marketing planningPlanning a year of ad spend without overspending not measured 13 of 18
Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls 152 of 161 77 of 161
Reasoning and knowledgeHard questions across science, maths and general knowledge 186 of 195 28 of 195
WritingWhat people prefer in blind comparisons, and judged writing quality 173 of 187 29 of 187
Sticking to the factsSummarising without inventing things, and factual answers 11 of 153 55 of 153
Speed under pressureGood decisions against a real clock (fast chess) not measured 22 of 22

Every shared result

BenchmarkMetricNova Pro 1.0DeepSeek V4 Pro 0423
Arena (formerly LMArena)reportedBusiness, management and financeArena rating 1,2791,460
Arena (formerly LMArena)reportedCreative writingArena rating 1,2371,446
Arena (formerly LMArena)reportedExpert promptsArena rating 1,2731,480
Arena (formerly LMArena)reportedInstruction followingArena rating 1,2771,452
Arena (formerly LMArena)reportedOverallArena rating 1,2901,458
Arena (formerly LMArena)reportedWriting, literature and languageArena rating 1,2741,451
Artificial AnalysisreportedArtificial Analysis Intelligence Indexindex score 7.0036.0
Artificial AnalysisreportedGPQA Diamond% of questions 49.9%92.8%
Artificial AnalysisreportedHumanity's Last Exam% of questions 3.2%41.0%
Artificial AnalysisreportedLong-context reasoning (AA-LCR)% of questions 21.0%80.3%
Vectara Hallucination LeaderboardreportedAnswer rate% of documents 99.3%97.2%
Vectara Hallucination LeaderboardreportedHallucination rate% of summaries 5.1%8.6%

Bold green marks the better value on that metric. Where a source reports ranges that overlap, the difference may not be meaningful; see the benchmark page for ranges.

Quick answers

Is Nova Pro 1.0 better than DeepSeek V4 Pro 0423?

It depends on the task. Among models available today, on our benchmarks and licensed sources, Nova Pro 1.0 ranks higher for sticking to the facts; DeepSeek V4 Pro 0423 ranks higher for agents and tool use, reasoning and knowledge and writing.

Which is cheaper, Nova Pro 1.0 or DeepSeek V4 Pro 0423?

Nova Pro 1.0 costs $1.40 per million tokens (three input to one output), against $1.98 for DeepSeek V4 Pro 0423.

Run Nova Pro 1.0 and DeepSeek V4 Pro 0423 on your own prompt

Benchmarks aren't your data. Try both side by side in the playground, with the cost of every answer.