Vibeleaderboard
← All Intel
Intel / post

We benchmarked DeepSeek V4 Flash, and it proved to be an absolute beast!

Source
x.com
Date
AlphaSignal@AlphaSignalAI

We benchmarked DeepSeek V4 Flash, and it proved to be an absolute beast! The results confirmed the claim that it is around 100× cheaper than Fable 5. Here’s what stood out: > @deepseek_ai V4 Flash ranked #1 at $0.004 per fix and $0.21 total > @grok 4.5 ranked #2 and was the fastest model with a perfect score at 55 seconds > @claudeai Fable 5 was the most expensive model with a perfect score at $0.389 per fix and $20.99 total How is this possible? Cheap cache reads. It actually processed more tokens than Fable 5, but most of its repeated context was billed at a discounted cache rate. This kept the total bill low even across millions of agent tokens. DeepSeek V4 Flash wasn’t the fastest model, but it delivered the strongest combination of debugging accuracy and cost. Full benchmark results and breakdown below.

View on X

Context

AlphaSignal's post and results chart report DeepSeek's V4 Flash completing 54 of 54 debugging fixes for $0.21 in total. Anthropic's Fable 5 also completed all 54, at $20.99, a roughly hundredfold cost gap.

AlphaSignal attributes the difference to cache pricing: DeepSeek processed more text overall, but repeated was billed at a reduced rate. The gap is therefore a result for this run and its billing conditions. It does not imply that unrelated workloads receive the same discount.

Terms in this piece · Glossary
  • context window — The maximum amount of text a model can consider at once — its working memory for the current conversation or task.
  • benchmark — A standard public test set for comparing AI models — the shared scoreboards behind every "model X beats model Y" claim.
  • token — The chunk of text a model reads and writes in — roughly three-quarters of a word — and the unit AI usage is billed in.
Read the source x.com
More from AlphaSignal
Recommended reads
Comments

Checking sign-in…

Loading comments…