Sign inGet Free $5

Changelog

Last updated: Oct 29, 2026

New models, new versions and price changes on OpenScale, newest first. Follow along with the RSS feed.

  1. New version

    Fast mode runs DeepSeek V4.1 Flash 1017

    deepseek/deepseek-v4.1-flash:fast now runs version 1017, the latest. The new pinned id deepseek/deepseek-v4.1-flash-1017:fast stays on 1017, and deepseek/deepseek-v4.1-flash-0910:fast stays on 0910. Prices are unchanged.

    Model catalog
  2. New model

    GLM 5.3 Flash is live

    Z.ai's fast multimodal model for coding and agent work, with a 1M-token context, text and image input, tool calling and reasoning. Model id z-ai/glm-5.3-flash, or z-ai/glm-5.3-flash-0828 to pin version 0828. Price: $0.12 input, $0.024 cached input and $0.40 output per 1M tokens.

    Model catalog
  3. New version

    Pinned model versions

    deepseek/deepseek-v4.1-flash always runs the latest version we have tested. To stay on one version, call a pinned id: deepseek/deepseek-v4.1-flash-1017 or deepseek/deepseek-v4.1-flash-0910. Pinned ids never change and cost the same.

    API reference
  4. New version

    DeepSeek V4.1 Flash 1017

    deepseek/deepseek-v4.1-flash moves to version 1017, published by DeepSeek on Oct 17, 2026. Same price and limits.

    Model catalog
  5. New model

    Fast mode for DeepSeek V4.1 Flash

    The same model on faster serving, at a higher price. Call deepseek/deepseek-v4.1-flash:fast, or send service_tier "priority" with the standard id. Price: $0.28 input, $0.028 cached input and $1.10 output per 1M tokens.

    Model catalog
  6. Price

    DeepSeek V4.1 Flash price cut in half

    Was $0.12 input, $0.006 cached input and $0.48 output per 1M tokens. Now $0.06 input, $0.003 cached input and $0.24 output per 1M tokens.

    Price comparison