Government Abuse is BAD for business. Getting ripped off on business phone / internet? Veteran-owned Business Telecom consulting. Free quotes USA lowest prices.
Wichita Kansas
Joined April 2009
- Tweets13.2K
- Following3.4K
- Followers988
- Likes16.5K
Pinned Tweet
#Developers and #VibeCoding without a spine is weakness!
I wrote a new article about AI where I propose some standards for agentic tooling. More to come soon:
linkedin.com/pulse/zippys-ad…
🤖 Made with AI
Ready to level up your gaming or local LLM AI SI rig? Me too... But I'm broke! I code for pizza and give away most everything I build. Buy 4, 2 for be and 2 for you and I'll give you 6k value of my time building stuff for you.
This is a pretty good deal right now at Walmart if you can find any in your area. I can go today and pick up several locally if anyone wants one let me know. Toss some change my way so I can get a pair too. My 4080 isn't enough for the demand I throw at it.
walmart.com/registry/TR/9c98…
What, no XO flying with all you buzz boys today? Do I need to call my Navy buddies to come show you how to fly straight?
@AFThunderbirds
Eric Henderson retweeted
NEW: Ukrainian President Volodymyr Zelenskyy tells @kilmeade he would accept a peace deal that freezes the war along the current frontlines, calling it a realistic way to stop the fighting.
He says Ukraine would then seek security guarantees from President Trump and European allies to ensure peace and security.
"We call it ceasefire or stop killing like President Trump called it. Freeze and stop. And of course this is minimum what we need from Russia today." | @foxandfriends
Eric Henderson retweeted
The only reason devs hate vibe coding is ego. They spent years feeling superior, and now a small local model writes their code.
Respect the craft, but fighting tech that shrinks years into days just to feel special is wild.
Adapt. Stop treating it like a threat. Your architecture and experience are what actually get the best out of these models. Shift your perspective.
What is going on in TN ? They apparently cannot execute a convicted death row killer. Reports of hours trying to execute Christa Pike. Governor stepped in and halted the execution. Sentenced to death at the age of 19. I'm just sad by all of this. Imagine finally thinking justice will be served only to see that. She lived. Imagine the same if you were being out to death by lethal injection but they screwed up or it didn't take.
Pentobarbital is used widely for this but something's off in this case. Governor ordered full investigation.
If death is the punishment, it should still be humane and swift and as painless for everyone as possible including the inmate.
I'm mixed about death punishment and especially for people that have not been given a chance for reform or working off the debt but I guess that's why there's strict times for when death penalty can even be on the table.
Make no mistake, this woman did awful things and was convicted and all appeals failed to stay the execution.
She should have died. Tennessee has some explaining to do. First hand witness accounts are all the same, "I've been to many executive... I've never seen something like this..."
Wow.
Holy Moly AI Power! Google dropped Gemini 4 blog.google/innovation-and-a…
Iniital reviews and tests are beyond impressive. Watch out OpenAI and Anthropic - when Google is on a misssion - they trample the competition.
We'll be compiling reviews and data over next week for full write-up.
Eric Henderson retweeted
We're giving away a CONTROL Resonant custom wrapped NVIDIA GeForce RTX 5080 to celebrate its official release with #RTXON.
To enter:
🟢 Share this post
🟢 Comment #RTXON
T&Cs: nvda.ws/4z24Gcq
Eric Henderson retweeted
Everyday this month @theo has tried to leave this room
Then something gets announced and he has to sit back down and talk about it
Eric Henderson retweeted
Hi,
Tomorrow we are re-opening the Pro $200 subscriptions to new subscribers, but together with it we are also changing how we calculate the usage for it. In effect, if you do the math, it will net out at half the dollar in API spend compared to the old Pro $200 plan.
Now that it's said, let me explain why this is happening and why you will still get more work done than if you were on the Pro $200 subscription one month ago.
(a) We didn't want to compromise in other ways and are committing to not reintroducing the 5h limit, so that you can fully use the weekly usage when you want.
(b) On the subscription, we guarantee that over time you always get more work done and with an increasing level of quality. This means that you will continue to get more value per dollar spent as a result of models getting more efficient and us passing down the improvements in the form of API price reductions.
(c) We don't want to put an incentive on ourselves to artificially inflate the API list prices to make it look like you are getting a lot (and workaround it through discounts, etc). Instead we want to continue to both rapidly reduce prices and increase capabilities of models on the API. This week we introduced GPT-6 Sol and GPT-6 Luna at 50% of their previous price. Over time, we see prices go low enough that it makes sense for most to buy usage as needed without there being a significant gap between what you get in a subscription and what you get in the API for a dollar spent.
(d) Tomorrow, we are adding more things to the subscription that won't draw on the usage, I won't reveal what that is yet.
I wanted to be transparent before all the big announcements tomorrow. Lots of new exciting things are coming to the subscriptions that will make it super compelling, but I wanted to make sure to share this change ahead of time so you can all understand it before we shower you with good news.
Codexingly,
Tibo
Q: Is Fable 5.1 on High the perfect sweet spot? I noticed a substantial improvement in quality and long orchestration intelligence for about the past 2 or 3 days and what was taking me days to accomplish with multiple adversarial review passes is just getting done now without any issue. I just let Fable decide what model to use for each task and offload it in subagents.
Eric Henderson retweeted
holy f*ck, Jev's founder leaked his favorite repos from github
10 fresh builds I didn’t use in the last list
voice-controlled browsers, Android notification filtering, intrusion detection, document splitting, live observability and semantic control flow
Voice Browser - controls a real Chromium window from partial speech; Jev decides the intent and target while you are still talking
↳ github.com/moritzkremb/jev-v…
DocJev - classifies and splits mixed PDF, DOCX and PPTX packets using natural-language document rules
↳ github.com/jerryjliu/docjev
Notiq - native Android notification filtering with plain-English rules; runs through Jev or a self-hosted FastJev endpoint
↳ github.com/chengyongru/notiq
Jev IDS - judges individual network flows as benign or malicious and routes attacks into fixed categories
↳ github.com/jev-ids/jev-ids
Jeview - a local proxy that records every Jev request in SQLite and draws the decision graph live
↳ github.com/andududu/jeview
JevLens - annotates articles, X posts and Threads posts using text plus image evidence extracted locally in Chrome
↳ github.com/knowlet/jevlens
Typeful Triage - a multiplayer dashboard that sorts GitHub issues and PRs by category, urgency, duplicates and the next maintainer action
↳ github.com/cephalization/jev…
Jev Reranker - reranks, filters or compresses JSON search results before they enter a RAG or agent context
↳ github.com/shinpr/jev-rerank…
jevc - compiles rules from files such as CLAUDE.md into narrow Jev questions and deterministic verdict code
↳ github.com/doronp/jevc
Discern - adds uncertainty-aware semantic pattern matching and control flow to Effect applications using Jev decisions
↳ github.com/doeixd/discern
pick by what you build:
> browser and accessibility → Voice Browser
> documents → DocJev
> Android → Notiq
> security → Jev IDS
> debugging and observability → Jeview
> social feeds → JevLens
> open-source maintenance → Typeful Triage
> search and RAG → Jev Reranker
> agent policy → jevc
> TypeScript infrastructure → Discern
the useful pattern across all ten is narrow:
Jev receives one bounded judgment
code still owns retrieval, permissions, thresholds, state and every side effect
these repos are young enough that a real bug report or integration PR can still change their direction
pick the one closest to your work, run the demo and break it
Eric Henderson retweeted
huihui-ai/Huihui-GLM-5.3-Flash-abliterated-Q8_0-GGUF
is now available on Ko-fi.
If you are interested, please consider making a donation through ko-fi.com/s/87851416fc to support our work.
We also accept BTC donations. Donation address: bc1qqnkhuchxw0zqjh2ku3lu4hq45hc6gy84uk70ge.
After donating, please send an email to support@huihui.ai telling us the name of the model you purchased and the last 8 characters of the transaction ID. We will then send you the download token.
If you need the non-quantized version or any other variants of huihui-ai/Huihui-GLM-5.3-Flash-abliterated, please feel free to send us an email to support@huihui.ai
New GGUF:
huihui-ai/Huihui-GLM-5.3-Flash-abliterated-GGUF
This is an uncensored version of zai-org/GLM-5.3-Flash created with abliteration.
Note
GGUFs come from unsloth/GLM-5.3-Flash-GGUF.
Only layers 15 to 35 (0-based indexing) have been ablated, while the other layers remain unablated.
huggingface.co/huihui-ai/GLM…
Eric Henderson retweeted
The open-source local version of Laya crushes Jev in response speed!
On the left is Laya, a local 421M open-source decision model.
On the right is Jev 1.13.0, running through a cloud API.
Same Snake game, same typed decisions, and the same 30-second free run:
- Laya: Score 46, Length 52, 86.5 decisions/sec
- Jev: Score 1, Length 7, 3.2 decisions/sec
- In a real test on an M5 Pro, median latency was 15.3ms locally versus 298.1ms in the cloud, so nearly 20x faster.
Laya uses around 1GB of memory, responds in milliseconds, and doesn’t depend on a network connection.
While Jev waits for a cloud round-trip at every step, Laya makes each decision instantly on-device.
Eric Henderson retweeted
Jev could become the control layer AI agents have been missing.
Instead of spending 5–20 seconds and expensive LLM calls deciding every next step, it can route actions in milliseconds at near-zero cost.
In this article, I break down how
@senecastreetliquor song and memes needed a new sound... suno.com/s/3nWC7WNt959vRjqm
WTH? This is INSANE! Have you heard anything new and fun in music lately? ... I heard something similar on FB and and I had to jump on board. I am so going to make an album for this AI group.
Zippy Yödelcore v1
suno.com/s/GW2Fwmz71m20W90j
Zippy Yödelcore v2
suno.com/s/vpOI9WMYTTCYMRiz
Which is better? Go ahead - troll away and toss some shade my way. This is fun!
