AI News

Trending story

Anthropic says AI ported Claude to AMD GPUs in three days

5 reports1 discussions6 story updatesUpdated Nov 12 17:56

Understand this story

AI overview

In investor materials, Anthropic said that when it first used AMD GPUs this year, the porting work was done by its frontier model Mythos, and Claude was running on AMD within three days with almost no human intervention, with the model writing nearly 100% of the kernels itself. Co-founder Tom Brown said the company has trained and run Claude on four major categories of chips, and treats compute from different vendors, architectures, and locations as an interchangeable resource pool; these figures come from investor presentation materials and have not been independently verified. On November 12, the blogger Dongxi Dongxi NLP raised doubts, saying Anthropic had become a "new short seller," arguing that by citing the undisclosed internal model Mythos to claim a three-day replication, it was attacking its competitors.

AI-generated from public reports

Latest development · Report

Blogger questions whether Anthropic is shorting via internal model Mythos

Developments

2 confirmed developments

  1. English text is being prepared.

    Source:English text is being prepared.
  2. Anthropic says it ported Claude to AMD GPUs in three days using AI

    Source:The Decoder

Reports and discussion timeline

Browse reports and discussion by time. Their evidence levels remain distinct.

Nov 12

  1. Ma Dongxi NLP

    Blogger questions whether Anthropic is shorting via internal model Mythos

    English text is being prepared.

  2. cnBeta

    Anthropic says Claude completed GPU kernel adaptation of a model to AMD accelerators in 3 days

    Anthropic says Claude handled almost all of the GPU kernel code porting and writing involved in adapting the model to AMD AI accelerators, with very little human intervention, and the entire process took only 3 days. According to the company's past experience, onboarding a new chip platform using traditional engineering methods typically requires dozens of engineers and takes 6 to 12 months. The chips actually adapted so far include NVIDIA, AMD, Google TPU, and Amazon Trainium. AMD and Anthropic previously announced they will deploy up to 2GW of GPU compute based on AMD's MI450 series, with the first 1GW planned to begin in the first half of 2027; the two sides will also use Claude to optimize AMD GPU workloads and accelerate ROCm software development.

  3. cnBetaRelated discussion

    cnBeta joined this discussion

    This post contributes to heat; it is not a formal report or confirmed fact.

  4. Phoenix Technology

    Huang Renxun Extramarital Affair Scandal Escalates as Nvidia Stock Plunges After Hours; OpenAI Posts $10.9 Billion Quarterly Loss | AI Morning Brief

    Nvidia's stock plunged more than 3% in after-hours trading on November 12, as CEO Huang Renxun was revealed to have had an extramarital affair with an internal executive who reported directly to him; the board had previously launched an independent investigation and the company faced a shareholder class-action lawsuit. OpenAI's third-quarter revenue grew about 50% quarter-over-quarter to $10 billion, but its net loss for the quarter reached $10.9 billion. xAI and SpaceX launched the Grok Voice 3.0 speech-to-speech API, priced at $0.08 per minute.

Nov 11

  1. Simon WillisonSelected

    Anthropic IPO roadshow video reveals multi-chip training strategy and approach to multiplying compute

    Anthropic has launched its IPO roadshow. In the roadshow video, co-founder Tom Brown introduced the company's compute strategy: Claude's training and inference now cover four types of chips: AWS Trainium, Google TPU, NVIDIA GPU, and AMD GPU. He mentioned that when first connecting to AMD GPUs this year, Claude completed the adaptation within three days, with almost no manual intervention, and ported or wrote nearly 100% of the kernels. Simon Willison used MacWhisper to extract the video transcript and organized the key points.

  2. The Decoder

    Anthropic says AI ported Claude to AMD GPUs in just three days

    Anthropic said in investor materials that when it first used AMD GPUs this year, its frontier model Mythos carried out the porting work, and Claude was running on AMD within three days with almost no human intervention, with the model writing nearly 100% of the kernels itself. Co-founder Tom Brown said the company has trained and run Claude on four major categories of chips, and treats compute from different vendors, architectures and locations as an interchangeable resource pool. These figures come from investor presentation materials and have not been independently verified.

Heat trend for this story

Current
5
Period peak
36 Nov 12 19:00
24-hour change
-75%

Move the pointer or select the chart to inspect hourly heat. Use the left and right arrow keys with a keyboard.

Related stories

  1. Related storiesAnthropic Publicly Submits S-1 for Proposed Nasdaq IPO
  2. Related storiesGoogle launches Gemini 4 Hydrogen on Antigravity
  3. Related storiesRecursive releases open-weight local coding model Flint
  4. Related storiesMeta launches Muse Wonder and Muse Wonder Full models