Bone Agent, Open Source - Software and agent enthusiast - Harnessmaxxer
Joined April 2026
- Tweets 1,177
- Following 102
- Followers 56
- Likes 271
17 Photos and videos
GLM 5.1 was not great at tui work, GLM 5.2 is much better. At the level of 5.5 or 4.8. I had some complex issues with getting scrolling live panes inside the terminal and it fixed the bug opus was stuck on.
Nice work @Zai_org
88
Rambone retweeted
Jun 13
BREAKING: The United States government has BANNED the export of three.js demo webpages
50
128
3,230
245,455
Jun 13
Not sure if im still rate limited for going over 700m tokens in the last month, but GLM 5.2 is still painfully slow. Wasn't like that 300m tokens ago. Still on the last few days of my legacy plan
230
Jun 13
I was going to cancel my glm plan bc I was getting rate limited before hitting my actual 5hr limits. But now I may keep it if this model is as good as the benches say. Glm 5.1 was not benchmaxed.
Jun 13
GLM-5.2 on KingBench (3).
Thoughts: The model has superb taste. It is greater at UX than UI. The code is always very clean. It is great at One-shot wonders. I asked it to fine-tune a whole local model and it did it in 30mins!
This is just a great model to use all-round.
1/n
225
Jun 13
Finally! Overdue. Was a huge fan of 5.1 cant wait to test out 5.2
Intelligence should be open, accessible, and ready to build with, empowering every developer, everywhere.
GLM-5.2 is now available to all GLM Coding Plan users, including Lite, Pro, Max, and Team plans.
docs.z.ai/devpack/latest-mod…
As our new flagship model, GLM-5.2 delivers powerful coding capabilities, usable 1M-context support, and continued strengths in long-horizon tasks.
API and Chatbot services will launch next week. The model will also be officially open-sourced next week under the MIT License.
The future of AI is open, and it belongs to the people.
23
Jun 13
I subbed again just to use fable. Obviously the news is bigger then my sub but im still sad.
Jun 13
The US government, citing national security authorities, has issued an export control directive to suspend all access to Fable 5 and Mythos 5 by any foreign national, whether inside or outside the United States, including foreign national Anthropic employees.
The net effect of this order is that we must abruptly disable Fable 5 and Mythos 5 for all our customers to ensure compliance.
Access to all other Claude models is not affected.
We apologize for this disruption to our customers. We believe this is a misunderstanding and are working to restore access as soon as possible.
Read our full statement: anthropic.com/news/fable-myt…
34
Jun 12
That was a fast release, and a really impressive bump. GLM where is 5.2?
Jun 12
Meet Kimi-K2.7-Code 👀
Here’s what developers should know to fully unlock K2.7-Code potential：
139
Rambone retweeted
Codex vs Antigravity vs Grok Build vs Claude Code
13 Dec 2025
Prompt: Create a 3D simulation of a formula 1 car performing a continuous drifting donut in the street using HTML/CSS/JS in a single HTML file.
16
9
194
56,335
Jun 10
I love chaining prompts. I can fire off a prompt then /new then another and repeat.
There's probably a better more complex way to do this but for quick tasks its nice.
1
1
28
Jun 10
We are about to hit a full week with no codex reset.
One must be due any minute now.
1
70
Jun 10
I recently started letting ai commit bc it writes way better commit messages then me. I actually know what each commit is now looking back.
Jun 10
TIL that lots of people have AI commit and push their changes.
Why would you do that?
Is it really that hard to run git commit & git push? You also save some tokens.
7
Jun 10
This is huge. Absolutely blazing speeds.
Qwen your move
Jun 10
Meet DiffusionGemma!
An experimental open model that explores a fast approach to text generation, released under an Apache 2.0 license.
Moving beyond sequential, token-by-token processes to generate entire blocks of text simultaneously. Here’s what’s new with DiffusionGemma: 👇
34
Jun 10
Why is GPT 5.5 leaking caveman speak into my chat. Is this a new feature?
"Recall already removed earlier? User says remove recall maybe stale docs. Need remove
protected? In current commands no recall.
Autocomplete maybe absent. Need update defaults/AGENTS? "
23
Jun 10
I bet fable is really good. I wouldn't know bc I hit the limit before it did anything.
Don't even try unless you are on the $100 plan.
1
15
Jun 10
I gave one extremely detailed prompt with exactly instruction on what to do to Fable on the $20 dollar plan. We are 60% through my 5hr limit and didnt even start writing code.
14
Jun 10
I took the bait and bought a claude pro plan to see what all this hype is about
9
Jun 9
Why are we still putting gemini on these benchmarks
14
Jun 9
I think i missed the original loops tweet bc I've only seen people talking about people talking about loops
3
Jun 9
This is going to be the nex unlock
1
14
Jun 7
Messing around with stats in the bone agent terminal. Tmux pop-up windows are so cool. For everyone else it uses the normal alt screen. Really liking the results:
1
24