GitSeonjeap
English

Sign in

Reading needs no sign-in. Sign in to write posts or comments, or to recommend and report, in the community.

By signing in you agree to the notices below. Community rules · Privacy notice

Keep browsing without signing in

← News

Industry

Meta Publishes Muse Ember via Meta Model API with Token Pricing Stated

Posted As of

Meta Superintelligence Labs said the Muse Ember model is immediately available through the Meta Model API and Muse Code. The blog lists prices of $0.10 per million input tokens and $0.50 per million output tokens.

What happened

  • Source: Meta Superintelligence Labs research blog
  • Muse Ember is offered as muse-ember-1.0 through the Meta Model API and Muse Code
  • Pricing: $0.10 per million input tokens, $0.50 per million output tokens (official)
  • The blog does not mention plans to release the weights
  • The blog's benchmark table has broken labels, so this article does not cover performance figures

Our take

The currently available routes are the API and Muse Code, so running this model locally on a Mac or PC directly would require the weights to be released. Meta has not said whether it will release the weights in the future. Actual work quality and cost should be checked separately against your own workloads.

Sources and related posts

More in Industry

  1. Anthropic Announces Claude Fable 5.2 and Fast Mode Available on Cloud Family (anthropic.com)

    Anthropic introduced Claude Fable 5.2 as its top model for technical work. The product page states that fast mode is available on Claude Platform, Google Cloud, and Microsoft Foundry.

    Industry| |1 sources |Our take included
  2. NVIDIA Releases GR00T 2 Robot World Action Model and Isaac Lab 3.0 (blogs.nvidia.com)

    NVIDIA introduced GR00T 2, an open model based on Cosmos, timed to CoRL. All performance figures are NVIDIA's own tests, and no information on the license or weights has been confirmed.

    Industry| |1 sources |Our take included
  3. Google Unveils Two New Systems, TPU 8tx and TPU 8i Max (blog.google)

    Google announced TPU 8tx, focused on security, and TPU 8i Max, built for large-context workloads. Reports differ on pricing and launch timing, so neither can be confirmed.

    Industry| |2 sources |Our take included

All news