DeepSeek-V5 Placeholders Spotted in Harness Codebase, Hinting at Next-Gen AI Model Rollout

On the evening of October 27, developers spotted a set of previously unseen DeepSeek-V5 model identifiers in the official DeepSeek Harness GitHub repository.

In the provider.spec.ts test file of the deepseek-harness repository, around line 154, three entries — deepseek-v5, deepseek-v5-open, and deepseek-v5-next — appeared alongside the existing deepseek-v4-pro. Based on the code structure, these entries currently act as framework-level placeholders rather than directly callable model interfaces. The v5-open entry includes requires('0.0.1'), indicating no substantial version gate, so older clients can continue working after a server-side switch; v5-next, however, specifies requires('99.0.0-alpha.1'), pointing to a channel that requires a client update. This suggests DeepSeek has prepared both "forced update" and "smooth transition" delivery paths for V5.

Earlier public information shows DeepSeek released the V4 series as a preview on April 24, 2026, with V4-Pro and V4-Flash; V4-Pro became generally available on August 13. In September, DeepSeek launched V4.1-Flash, which was described as using a new architecture but remained under the V4 naming.

As of October 28, DeepSeek had not made any official announcement, model card, or API identifier for the V5 series, nor publicly commented on the placeholders.