DeepSeek V3
by DeepSeek
DeepSeek-V3 is the latest model from the DeepSeek team, building upon the instruction following and coding abilities of the previous versions. Pre-trained on nearly 15 trillion tok
Models
DeepSeek V3
Added 10 July 2026
Overview
DeepSeek-V3 is the latest model from the DeepSeek team, building upon the instruction following and coding abilities of the previous versions. Pre-trained on nearly 15 trillion tokens, the reported evaluations...
Best for
Best for
High-volume cheap inference and drafts
Use cases
- classification
- draft generation
- bulk enrichment
How to use / API access
Call via OpenRouter with model id `deepseek/deepseek-chat`.
- openrouter ·
deepseek/deepseek-chatDocs
OpenRouter id: deepseek/deepseek-chat
Benchmarks
- Design Arena 3d (other): 1087 elo · #90
- Design Arena codecategories (other): 1110 elo · #100
- Design Arena dataviz (other): 1090 elo · #101
- Design Arena gamedev (other): 1055 elo · #107
Enterprise DNA angle
Route bulk Omni jobs to cheap DeepSeek rows; escalate only failures to frontier models.
Notes
DeepSeek V3 is listed in the EDNA Models directory from the OpenRouter catalogue.
Pros
- Low $/Mtoken for high volume
- Solid general chat and light coding
Cons
- Not the top pick for hardest reasoning
- Needs eval gates for brand-sensitive copy
Get the free Developer’s Field Guide
A 27-page field guide to the AI coding workflow with Claude. Claude Code, MCP servers, the prompt patterns that work, and what to delegate. Free.
Enter your work email. We send it straight over, plus a few short notes worth knowing. Unsubscribe any time.