PulseAugur
EN
LIVE 04:37:17
中文(ZH) 范式下场做基模了!国产算力全链路训练,双模型同步亮相

Phanthy AI launches Phai-A1 language and Phai-M1 video models

Fourth Paradigm (now Phanthy AI) has launched two new AI models: Phai-A1, a general-purpose language model, and Phai-M1, a video generation model for realistic film and television scenarios. Phai-A1 boasts an 8-10x faster response time than comparable models, can run on a single consumer GPU like the RTX 3090, and supports a native million-token context window. Phai-M1 generates 15-second 720p videos in approximately 3 minutes with synchronized audio and video, aiming for cinematic quality. Both models were trained end-to-end on domestic Chinese computing infrastructure, specifically Huawei Ascend processors, and are slated for release on Hugging Face and ModelScope. AI

IMPACT These models, particularly Phai-A1's efficiency and context window, could influence the development of smaller, more accessible LLMs and specialized video generation tools.

RANK_REASON New model release from a notable AI lab. [lever_c_demoted from frontier_release: ic=1 ai=1.0]

Read on 量子位 (QbitAI) →

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →

Phanthy AI launches Phai-A1 language and Phai-M1 video models

How we ranked this

Signal score
57 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Significant
New model release from a notable AI lab. [lever_c_demoted from frontier_release: ic=1 ai=1.0]
Source corroboration
Single-source cluster
Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics
model release, infra
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
Breaking (< 6h)
Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

Full methodology in our editorial standards.

COVERAGE [1]

  1. 量子位 (QbitAI) TIER_1 中文(ZH) · 思邈 ·

    Paradigm enters the field to do base models! Domestic computing power full-link training, dual models debut simultaneously

    语言模型单张3090卡即可运行