Open Source
DeepSeek V4 Offers Cost-Effective AI Solution
Matt Wolfehigh confidence
Why it matters
- →DeepSeek V4 offers a cost-effective alternative to expensive AI models.
- →Its open-source nature allows for greater accessibility and customization.
- →Running locally enhances privacy, appealing to privacy-conscious users.
DeepSeek V4 has been introduced as a nearly state-of-the-art AI model that is open-sourced and offers substantial cost savings. It charges $1.74 per million input tokens, which is significantly cheaper than the $5 charged by competitors like GPT-5.5 and Claude Opus 4.7. Additionally, DeepSeek V4 can be run locally, providing users with enhanced privacy. This combination of affordability, capability, and privacy could potentially disrupt the current AI market landscape.
Read originalThe story around this
Earlier coverage that leads up to this article, and what followed. Lines connect each piece to the closest one after it, converging here.
How we got here
More from Matt Wolfe
More in Open Source
© Lev SelectorOpen Sourcemodels
Bonsai 2 and Needle: Tiny Local AI Models
New tiny local models Bonsai 2 and Needle (8-29 MB) demonstrate that small, offline-capable AI can make fast, useful decisions.
Lev Selector
Open Sourceother
llama.cpp b11153 adds ROCm 10 and Snapdragon support
This release quietly cements llama.cpp as the universal inference runtime by finally bringing first-class ROCm 10.0 support to both Linux and Windows. AMD GPU users no longer need workarounds, effectively closing a long-standing parity gap with NVIDIA's CUDA ecosystem. Equally notable is the new Linux arm64 build targeting Snapdragon chips, which unlocks local AI on high-performance mobile hardware via CPU, Adreno GPU, and Hexagon NPU acceleration. While KleidiAI on Apple Silicon has been disabled in this specific binary, the broader platform expansion signals that llama.cpp is aggressively standardizing how models run across every major silicon architecture.