Glonce

DeepSeek V4 Pro 0813 reaches GA with 1M context, $0.435/$0.87 pricing

modelsThe Glonce editorsHacker News🔥68Glonce rating68Impact45Novelty38Relevance95Surprise22Players40Credibility80
DeepSeek V4 Pro 0813 reaches GA with 1M context, $0.435/$0.87 pricing

DeepSeek V4 Pro 0813 is now listed on OpenRouter as the general availability (GA) release of DeepSeek V4 Pro, described on the listing as a large-scale mixture-of-experts model. The listing gives a release date of August 12, 2026. OpenRouter shows DeepSeek as the single provider hosting the model, so requests are forwarded directly to DeepSeek with no routing decision to make. The model supports a 1M-token context window. Pricing is $0.435 per million input tokens and $0.87 per million output tokens. The model is reachable through OpenRouter's OpenAI-compatible API, meaning most existing SDKs work by swapping the base URL and model slug rather than requiring new integration code. Beyond the GA announcement, pricing, context length and hosting details, the listing does not state what changed relative to earlier DeepSeek V4 releases, does not show benchmark scores or performance figures, and carries no comparison to competing models or any statement from DeepSeek about the release.

Key facts

  • DeepSeek V4 Pro 0813 is the GA release of DeepSeek V4 Pro, a mixture-of-experts model, dated August 12, 2026.
  • It is listed on OpenRouter with DeepSeek as the sole hosting provider, forwarding requests directly with no routing step.
  • Pricing is $0.435 per million input tokens and $0.87 per million output tokens.
  • The model has a 1M-token context window and is reachable through OpenRouter's OpenAI-compatible API.
  • The listing shows no benchmark scores, no comparison to prior DeepSeek versions or rivals, and no statement from DeepSeek about what changed; even so, the Hacker News discussion drew 825 points and 326 comments within about 13 hours.

Why it matters

A GA listing for DeepSeek V4 Pro 0813 puts another large mixture-of-experts model with a 1M-token context window into general availability at sub-dollar per-million-token pricing. That combination of context length and price is what OpenRouter's listing highlights, even though the page itself offers no benchmark numbers to back up how the model performs.

Who it affects

Developers and companies building on LLM APIs, particularly those already routing traffic through OpenRouter or comparing providers on price and context length. Anyone evaluating DeepSeek's lineup for long-context workloads is the direct audience.

How to use it

The model is available now through OpenRouter, with DeepSeek as the only provider hosting it, so there is no routing decision between multiple hosts. Access goes through OpenRouter's OpenAI-compatible API: input tokens cost $0.435 per million and output tokens cost $0.87 per million, and the context window is 1M tokens.

How solid is it

The listing itself is the only source available: model name, provider, release date, pricing and context length are all stated directly on the OpenRouter page. What is missing is any benchmark score, any comparison to earlier DeepSeek V4 releases or to competing models, and any quoted statement from DeepSeek about the release. The strong Hacker News response, 825 points and 326 comments in roughly 13 hours, reflects interest in the release, not independent confirmation of the model's capabilities.

Risks and caveats

Without benchmark data or a stated comparison to prior versions, there is no way from this listing to judge whether DeepSeek V4 Pro 0813 represents a capability improvement over earlier DeepSeek models or how it stacks up against competitors at similar pricing. The pricing and context figures come from the listing itself and could change as OpenRouter updates provider terms.