msp26

7 ms·

31.
by msp26 7mo ago
Apologies but I will use this thread as an opportunity to report CC VSCode extension bugs because I don't think there's an official channel that actually gets read by humans. > yeah they're shipping too fast and everything
32.
by msp26 7mo ago
many tasks don't need any reasoning
33.
by msp26 7mo ago
What the fuck is this price hike? It was such a nice low end, fast model. Who needs 10 years of reasoning on this model size?? I'm gonna switch some workflows to qwen3.5. There's a lot of tasks that benefit from just having a mild
34.
by msp26 7mo ago
> every single product/feature I've used other than the Claude Code CLI has been terrible yeah they're shipping too fast and everything is buggy as shit - fork conversation button doesn't even work anymore in vscode e
35.
by msp26 7mo ago
Batshit situation, respectable position from Dario throughout. But there's some irony in this happening to Anthropic after all the constant hawkish fearmongering about the evil Chinese (and open source AI sentiment too).
36.
by msp26 8mo ago
Horrific comparison point. LLM inference is way more expensive locally for single users than running batch inference at scale in a datacenter on actual GPUs/TPUs.
37.
Tool Shaped Objects(minutes.substack.com)
3 points by msp26 8mo ago|0 comments
38.
by msp26 8mo ago
https://minutes.substack.com/p/tool-shaped-objects I feel like this applies for many of you.
39.
by msp26 8mo ago
> Special shout out to Google who to this date seem to not support tool call streaming which is extremely Google. Google doesn't even provide a tokenizer to count tokens locally. The results of this stupidity can be seen directly in
40.
by msp26 8mo ago
This doesn't surprise me. I have a SKILL.md for marimo notebooks with instructions in the frontmatter to always read it before working with marimo files. But half the time Claude Code still doesn't invoke it even with me mentionin
41.
by msp26 8mo ago
Source? I've heard this rumour twice but never seen proof. I assume it would be based on tokeniser quirks?
42.
by msp26 8mo ago
K2 thinking didn't have vision which was a big drawback for my projects.
43.
by msp26 9mo ago
Thank you! That looks great.
44.
by msp26 9mo ago
Mildly related question for the people in the thread: How do I seek to the exact first frame of a timestamp with mux? I've tried a few things but it seems to always go to the nearest keyframe rather than the first frame at e.g. 00:34.
45.
by msp26 9mo ago
Originally I thought that Gas Town was some form of high level satire like GOODY-2 but it seems that some of you people have actually lost the plot. Ralph loops are also stupid because they don't make use of kv cache properly. --- htt
46.
by msp26 9mo ago
This account's comment history is pure slop. 90% sure its all AI generated. The structure is too blatant.
47.
by msp26 9mo ago
Incredible guide, wow. Will definitely share with people. I wish I had something like this a year ago.
48.
by msp26 9mo ago
> because there's already concern that AI models are getting worse. The models are being fed on their own AI slop and synthetic data in an error-magnifying doom-loop known as "model collapse." Model collapse is a meme that
49.
by msp26 10mo ago
Hi if the Gemini API team is reading this can you please be more transparent about 'The specified schema produces a constraint that has too many states for serving. ...' when using Structured Outputs. I assume it has something to
50.
by msp26 10mo ago
The new large model uses DeepseekV2 architecture. 0 mention on the page lol. It's a good thing that open source models use the best arch available. K2 does the same but at least mentions "Kimi K2 was designed to further scale up M
51.
by msp26 10mo ago
K2 Thinking has immaculate vibes. Minimal sycophancy and a pleasant writing style while being occasionally funny. If it had vision and was better on long context I'd use it so much more.
52.
by msp26 10mo ago
Because its not a software issue, it's a human social cooperation issue. Companies don't want to support useful APIs for interoperability so its just easier to have an LLM bruteforce problems using the same interface that humans u
53.
by msp26 11mo ago
really nice post, will share!
54.
by msp26 11mo ago
Is flash/flash lite releasing alongside pro? Those two tiers have been incredible for the price since 2.0, absolute workhorses. Can't wait for 3.0.
55.
by msp26 11mo ago
https://saucenao.blogspot.com/2021/04/recent-events.html Mildly related incident where a Canadian child protection agency uploads csam onto a reverse image search engine and then reports the site for the temporari
56.
by msp26 11mo ago
> I don't like how closed the frontier US models are, and I hope the Chinese kick our asses. For imagegen, agreed. But for textgen, Kimi K2 thinking is by far the best chat model at the moment from my experience so far. Not even &qu
57.
by msp26 11mo ago
Groq does quantise. Look at this benchmark from moonshotai for K2 where they compare their official implementation to third party providers. https://github.com/MoonshotAI/K2-Vendor-Verifier It's one of the lowest
58.
by msp26 1y ago
Rumour is a release on the 22nd I believe
59.
by msp26 1y ago
Accessing services from the UK without handing over your personal ID to a service that will inevitably get hacked. This happened to discord literally a few days ago.
60.
by msp26 1y ago
The voice quality in the generated vids is surprisingly awful.

More ›