Tokenstead - Open AI Models, Tooling & Hardware
New - local AI
Claude will watermark AI-generated text and images
Anthropic says new Claude models launched in the EU on or after August 2, 2026 will embed text watermarks and C2PA provenance metadata on supported files. The marks help with transparency, but can be lost through editing, screenshots, or format conversion.
How Claude marks AI-generated content
New - local AI
Cloudflare OS: an open platform for agents, apps, and work
Just announced and #1 on Hacker News. Cloudflare OS is a browser-based agent workspace where non-developers can build apps, workflows, and automations grounded in company context, with audited capability-based access to internal systems.
Read on Cloudflare Blog
New - local AI
Kimi K3 now runs locally - Unsloth's 1-bit quant shrunk it 62%
The strongest open model to date (2.8T MoE) fits in RAM on a 748GB DGX Station at 594GB (-62% from 1.56TB), ~78.9% accuracy retained. On a Mac, it loads via SSD memory-mapping with a ~128GB working set - slower, but it runs.
See the model card
Run AI on hardware you own - and follow the models worth running
Track the latest open-weight models, agent harnesses, and autonomous agents, then see which ones fit your rig. Honest speed estimates, cloud-pricing comparisons, and a source-cited tracker of who's running what - so no vendor or government order can switch off the model you depend on.
Browse the latest models<br>→
Find models for your hardware<br>↓
Latest open-weight models
Browse all models
01
Muse Glimmer 30B<br>enthusiast
30.0B params - 128k ctx<br>- released Aug 2026
→
02
Muse Spark 1.2<br>premier
0.0B params - 1000k ctx<br>- released Aug 2026
→
03
Pokee-Isaac 28B<br>enthusiast
28.0B params - 10000k ctx<br>- released Aug 2026
→
04
Qwen3.8-Max<br>MoE<br>premier
2400.0B params - 1000k ctx<br>- released Aug 2026
→
05
DeepSeek V4 Flash 0731<br>MoE<br>workstation
284.0B params - 1000k ctx<br>- released Jul 2026
→
06
Kimi K3<br>MoE<br>premier
2800.0B params - 1000k ctx<br>- released Jul 2026
→
Latest AI tooling
Agent harnesses<br>Autonomous agents<br>Business AI<br>Marketing agents<br>Agent workspaces
OpenClaw<br>Autonomous agents
Local-first personal AI assistant across 25+ messaging channels
★ 385,724
Hermes Agent<br>Autonomous agents
Self-improving AI agent that learns and grows with you
★ 228,046
MIT
n8n<br>Marketing agents
Open-source workflow engine with AI agent nodes for the decision loop
★ 200,029
Sustainable Use License
OpenCode<br>Agent harnesses
AI coding agent for the terminal, desktop, and IDE
★ 195,530
MIT
Codex<br>Agent harnesses
Lightweight coding agent that runs in your terminal
★ 105,016
Apache-2.0
Latest news & guides
All guides
Cloudflare Wallets and x402: agents that pay for inference
Aug 5, 2026
How to Spend $200/Month on AI in 2026
Jul 29, 2026
Opus 5 ultracode wiped a production database: postmortem
Jul 29, 2026
How to Build a Marketing Agent With Open-Source Tools
Jul 28, 2026
Run the Grok CLI on Ollama Cloud and custom providers
Jul 19, 2026
Top models by quality
Browse all models
01
Qwen3.8-Max<br>MoE
2400.0B params - 1000k ctx
91<br>GENERAL
02
DeepSeek V4 Flash 0731<br>MoE
284.0B params - 1000k ctx
90<br>GENERAL
03
Kimi K3<br>MoE
2800.0B params - 1000k ctx
90<br>GENERAL
04
DeepSeek V4 Pro<br>MoE
1600.0B params - 1000k ctx
89<br>GENERAL
05
GLM 5.2<br>MoE
744.0B params - 1000k ctx
88<br>GENERAL
06
Pokee-Isaac 28B
28.0B params - 10000k ctx
88<br>GENERAL
Find the right model for your hardware
Already know your rig? Pick it here and see exactly which models you can run locally, with honest speed estimates and cloud-pricing comparisons.
01 - Select your hardware
click to compare
AMD Ryzen AI Halo 128GB
128GB UNIFIED
Framework Desktop 128GB
128GB UNIFIED
GMKtec EVO-X2 128GB
128GB UNIFIED
Mac Mini M4 16GB
16GB UNIFIED
Mac Mini M4 Pro 24GB
24GB UNIFIED
Mac Mini M4 Pro 48GB
48GB UNIFIED
Mac Studio M4 Max 36GB
36GB UNIFIED
Mac Studio M4 Max 64GB
64GB UNIFIED
Mac Studio M4 Max 96GB
96GB UNIFIED
or enter custom specs
On a budget? Compare own vs rent →
Count tokens & estimate API cost →
Or have us run a managed private AI rig →
Unified (Mac/Jetson)<br>VRAM (GPU card)<br>System RAM
02 - Save your rig (free)
Sign in with GitHub to save your hardware. New here? We'll guide you through picking your rig - Mac, multi-GPU (up to 8x), or custom specs - then show you exactly which models you can run locally.
Sign in with GitHub - free
Who's running what
See all 23 →
Microsoft<br>runs<br>Kimi K3<br>testing<br>Moonshot AI
The Information: engineers evaluating Moonshot AI's Kimi K3 (2.8T open-weight, released 2026-07-16, $3/$15 per MTok) for Copilot features currently on GPT/Claude, citing strong coding benchmarks and ~60% lower inference cost. Not officially confirmed; evaluating, not deployed.
Microsoft<br>runs<br>MAI<br>reported<br>Microsoft
Bloomberg: Microsoft replacing OpenAI/Anthropic models with in-house MAI models in Excel and...