The Information Machine
Concluded·following since 6 Aug 2026·Day 3·2 sources·updated 8 Aug 2026

DeepSeek API price hike

The gist

DeepSeek announces significant API price hike as explosive demand made loss-leader pricing unsustainable

DeepSeek's current pricing is roughly 105x cheaper per task than Claude Fable 5, a gap that drove rapid adoption and set a cost benchmark in the market. A significant price increase reduces that advantage at the same moment rivals have closed the capability and price gap.

The full picture

DeepSeek announced its API prices will rise 'significantly' from current rates of 14 cents/28 cents per 1M input/output tokens, and told users to plan accordingly. The increase follows the release of DeepSeek-V4-Flash-0731, an open-source model that nearly matches Anthropic's Opus 4.8 in capabilities at a fraction of the cost. The model became the fastest-growing ever by token usage on Ollama, prompting Ollama to expand capacity across the US and Europe. The price hike comes as Meta's Muse Spark and OpenAI's GPT-5.6 Luna now match DeepSeek on both capability and price. Among open-weight models, DeepSeek-V4-Flash-0731 ranks second, behind Moonshot AI's Kimi K3. DeepSeek also resumed its $8 billion funding round, previously paused after a CEO comment leak, at a $74 billion valuation.

How it developed
8 August 2026

DeepSeek announced August 7 that API prices will rise significantly from 14 cents and 28 cents per million input and output tokens, driven by demand from its open-source DeepSeek-V4-Flash-0731, which became the fastest-growing model by token usage on Ollama and nearly matches Anthropic's Opus 4.8 at a fraction of the cost.

Meta's Muse Spark and OpenAI's GPT-5.6 Luna now match DeepSeek on both capability and price. DeepSeek also resumed its $8 billion funding round, paused after a CEO comment leak, at a $74 billion valuation.

7 August 2026

DeepSeek resumes $8 billion funding round at $74 billion valuation after CEO comment leak pause

6 August 2026

DeepSeek announces API pricing will rise 'significantly'; V4-Flash-0731 described as fastest-growing model ever by token usage on Ollama

Sources
The daily email

Want this in your inbox?

I send a short email each morning with the stories that moved. If you would rather just read here, that works too.

Subscribe free