DeepSeek-V4-Flash-0731 launches with agent upgrades

DeepSeek released V4-Flash-0731, a 304B MoE model, with Terminal-Bench 2.1 jumping to 82.7 from 72.1. It costs $0.10 per DeepSWE rollout vs GPT-5.6 Luna's $0.61, delivering 4.8x solves per dollar.
How this story unfolded
4 weeks · 10 reports · 56 community posts · 66 of 70 shown
- Jul 26
- Jul 31
- Aug 1
- Aug 2
- Aug 3
- Aug 4
- Aug 5
- Aug 6
- Aug 7
- Aug 8
- Aug 9
- Aug 10
- Aug 12
- Aug 13
- Aug 14
- Aug 18
- Aug 19
- Aug 20
DeepSeek by email
Get an email when DeepSeek has news
No news that day, no email.
More stories today
- Apple applies iterative pseudo-labeling to code-switching ASR
- Vercel Agent is now available in Slack code channels
- Doctorow: AI's epistemic crisis is an 'opportunistic infection'
- Gary Marcus: OpenAI is becoming a surveillance company
- agtx runs multi-agent coding workflows from a kanban board