Liquid AI open-sources on-device benchmark Pipette
It measures speed and memory by model-quantization-runtime-device combination, and ships with over 10,000 verified results
20
AMD
A chip company often seen as NVIDIA's main rival. It makes the "Instinct" line of AI accelerators. It also makes the "EPYC" line of server processors. In August 2026, AMD unveiled new EPYC chips while lowering prices. That same month, the open-source project llama.cpp added a CI build target aimed at ROCm 7.14, AMD's GPU software stack.
Current rank (1M)
27
- Change
- Prev.
- 23
Rank over the last 30 days
2026.08.18 – 2026.09.16 · High 15 · Low 27 · Now 27
AMD · Related stories
Waymo Develops Its Own Robotaxi Chip, Cutting Reliance on NVIDIA
The chip, built on TSMC's 5nm process, delivers 1,000 TOPS of performance, matching NVIDIA's self-driving systems
30
Inherent's 27B Model Faraday Outperforms Anthropic and OpenAI
A London startup founded by Google DeepMind alumni beat frontier models at a paper-replication task using a much smaller model
30
Modular, Now Owned by Qualcomm, Fully Open-Sources Mojo Language
Compiler included under Apache 2.0 — MiniMax already serving billions of tokens per minute
60
US Rate Hikes Rattle AI Rally as AMD Unveils New EPYC, GPT-5.6 Gets Price Cut
As surging Treasury yields threaten Asia's AI stock rally, AMD's next-generation server chip, a GPT-5.6 price cut, and open-source model and infrastructure discussions filled the day.
10
Liquid AI unveils screen-reading model that runs in 3GB
LFM2.5-VL-3B hits 20 tokens/sec on Galaxy S26 Ultra, 228 tokens/sec on M5 Max
00
Anthropic Signs $9.1 Billion Data Center Deal With Bitcoin Miner
Deal secures 191 megawatts at Riot Platforms' Texas site, could grow to $16.1 billion with extensions
00
llama.cpp adds CI build target for AMD ROCm 7.14
First official release built on TheRock build system expands Linux and Windows support
170
Tiny Corp Runs 27B Model at 34 Tokens per Second via USB3-Connected GPU
Tiny corp released a video showing a 7900XTX connected over USB3 running Qwen 3.6 27B. The dock ships on the 12th
40
Meta releases 30B-parameter open model for local agents
Muse Glimmer is an agent-specialized model that runs on 24GB-class consumer GPUs with 4-bit quantization
180
audio.cpp 0.5 released, adds expressive TTS and cross-lingual voice transfer
Seven new models added, plus ROCm/HIP support. The local voice stack keeps getting thicker.
10
That's the last story.
AMD · Recent announcements
- 597Meet ROCm 10. Here are 10 things #AMDevs need to know.
- 222The future of AI will be built by people with access to open technology, skills and opportunity.
- 237From vision to production. AI infrastructure powered by AMD Instinct MI355X GPUs and AMD EPYC CPUs is now live in Saudi Arabia and serving @HUMAIN customers.
- 238#IYKYK: Different workloads demand different kinds of performance.
- 140#IYKYK: Agentic performance is an end-to-end systems equation.
- 137Save the date for IFA 2026! @jackhuynh, AMD SVP & GM of Computing and Graphics, will deliver the opening keynote and share the AMD vision for the next era of personal AI.
- 111AI is moving into the physical world.
- 85More throughput. Lower token economics. Greater deployment flexibility.
- 78Agentic AI is accelerating creative workflows.
- 79Creating richer digital worlds takes more than imagination. It takes breakthrough compute.
- 61Good luck to Kimi Antonelli, @GeorgeRussell63, and the entire @MercedesAMGF1 team as they head into the Italian Grand Prix.