AI is changing the world. Don't stay behind. Clear summaries, community insight, delivered without the noise. Subscribe to never miss a beat.

DeepSeek V4 Pro at 5% the cost of Claude – what it takes to close the gap

howardchen.substack.com

June 16, 2026

12 min read

🔥🔥🔥🔥🔥

44/100

Summary

DeepSeek V4 Pro is available at 5% of the cost of Claude and offers a range of features including hash-anchored edits, a sticky prefix cache, and autonomous loops for production code. It has been used for various applications such as training dose-prediction models for radiotherapy and developing a financial research agent, with no associated charges.

Key Takeaways

DeepSeek V4 Pro costs $0.435 per million input tokens, making it 5-7 times cheaper than Claude Sonnet 4, which costs around $3 per million tokens.
DeepSeek V4 Pro performs at approximately 80-85% of Claude's capability on long tasks, but achieves around 90% effectiveness in practical workflows with custom harness adjustments.
DeepSeek V4 Pro excels in following precise specifications and producing correct numerical and scientific code, while struggling with long-horizon planning and reading poorly structured code.
The implementation of hash-anchored editing significantly improved the efficiency of DeepSeek V4 Pro, reducing output tokens by 61% and enhancing performance on coding benchmarks.

Read original article