how to run them
FREE / local
━━━━━━━━━━━━━━━━
Qwen 3.8 27B Heretic Abliterated
GGUF → huggingface.co/0bserverx/Qwe…
Run in llama.cpp / LM Studio → plug into OpenCode, Cline, or Cursor
────────
Qwen 3.8 Flash Next Abliterated
HF → huggingface.co/huihui-ai/Hui…
GGUF → huggingface.co/huihui-ai/Hui…
(large — needs VRAM or cloud)
────────
Dolphin 3.0 Mistral 24B
HF → huggingface.co/dphn/Dolphin3…
Ollama → ollama run Jotschi/dolphin-mistral-24b
━━━━━━━━━━━━━━━━
PAID · Venice (not free on OpenRouter)
━━━━━━━━━━━━━━━━
Venice Uncensored 1.2
venice.ai/models/venice-unce…
~$0.20 / $0.90 per 1M
────────
Gemma 4 Uncensored
venice.ai/models/gemma-4-unc…
────────
DeepSeek V4.1 Flash Uncensored
venice.ai/models/deepseek-v4…
────────
GLM 4.7 Flash Heretic (cheapest)
venice.ai/models/olafangensa…
~$0.07 / $0.40 per 1M
────────
9 free MIT agent skills for Claude Code and Codex 😳
Worth knowing if you ship with AI coding tools.
Two of them sound almost identical the difference matters:
→ Full Send takes a change all the way to a merged PR with nobody watching
→ Full Stop does the same work, then stops with the PR ready for your review
Learning or a shared repo → start with Full Stop
Your own sandbox + solid checks → Full Send is the fast path
One rule: open every SKILL.md before you copy it in
A public repo is readable.
That is not a security audit.
Skill files are free. Claude Code, Codex and any model calls are on you.
quick note morphllm.com/students is 404 for a lot of people rn
if it opens, SheerID + school email, no card
if not, wait it out and check with vp..
DeepSeek V4.1 Flash is FREE on Freebuff
no card and no top-up need 😁
→ free coding models on desktop, phone, and web
→ ads pay for it so you stay at $0
→ daily hours by region, not unlimited
→ still enough for real daily build work
try it: freebuff.com
FREE coding API with 1M context and no daily request cap 😳
MiniMax M3.1 Flash on AnyRouter is $0 right now
OpenRouter still bills the same model
anyrouter.dev
Many creators are getting millions of impressions just by posting
“How I got 1M or 5M impressions using Grok bots”
😂Most of their stories might be made up but the impressions they’re getting from these guides are REAL.
DeepSeek V4.1 Flash is free every night 😳
On WhatsToken.
10 PM to 8 AM (UTC+8)
No card needed
whatstoken.ai
IMO DeepSeek V4.1 Flash Uncensored is almost FREE 😳
Setup (30 sec):
1. open larprouter.com
2. login with GitHub or Google
3. select DeepSeek V4.1 Flash Abliterated FP8
includes:
- no filters on DeepSeek V4.1 Flash
- 1M token context
- zero retention
- $0.12 / 1M tokens
Fable 5.5 might be coming sooooon 👀
Now I am curious what OpenAI is cooking up to counter it.....
Anthropic has filed an S-1 registration statement with the SEC related to its proposed IPO.
Read more: anthropic.com/news/s1-sec
Legal disclaimer: A registration statement relating to these securities has been filed with the SEC but has not yet become effective. These securities may not be sold, nor may offers to buy be accepted, prior to the time the registration statement becomes effective. This post does not constitute an offer to sell or the solicitation of an offer to buy these securities, nor shall there be any sale of these securities in any state or jurisdiction in which such offer, solicitation or sale would be unlawful prior to registration or qualification under the securities laws of any such state or jurisdiction.
