iAkashPaul

6 ms·

31.
by iAkashPaul 2y ago
Realtime translation of text in signage & other things is going to be one of the killer apps for pass-thru AR!
32.
by iAkashPaul 2y ago
Is this applicable to public & private buckets alike? Like if I keep making requests with the wrong token?
33.
by iAkashPaul 2y ago
My go to for lightweight deployments is HuggingFace's Spaces, free 2-core CPU, anything else can go to either serverless container images on runpod or baseten.
34.
by iAkashPaul 2y ago
With unsloth's optimizations you can do llama-3-8b's QLoRA fine-tuning on your 8GB card(mine's a 2070S) with 900MB to spare with BS of 4.
35.
by iAkashPaul 2y ago
Slip in a line asking them to reverse the string - .DefaultCellStyle
36.
by iAkashPaul 2y ago
I actually want Play Store apps to be equally configurable. Apps like Truecaller refuse to work without so many permissions. Being Android you'd expect some way to provide a dummy contact book, location & other accesses that apps n
37.
by iAkashPaul 2y ago
Seems like using RAFT( https://techcommunity.microsoft.com/t5/ai-ai-platform-blog/r... ) to generate the glossary+KGs for a given doc should give better results for cross-doc or intra doc answers instead of just rel
38.
by iAkashPaul 2y ago
Yep, termux is a good way to do this. Llama.cpp has Android example as well, I forked it here GitHub.com/iakashpaul/portal you can try it with any supported GGUF/Q4+Q8 models
39.
by iAkashPaul 2y ago
You can either modify the Android example inside llama.cpp or my fork of it at GitHub.com/iakashpaul/portal Increase the Ctx to more than 100 & link to any Q4 GGUF of 7B
40.
by iAkashPaul 2y ago
You can plug sileroVAD in the browser for this sort of interruptions, if you can make use of threads/workers in JS then you can mute/stop your output & instead have the chunks dumped to the STT websocket
41.
by iAkashPaul 3y ago
It was also possible to install an OS without needing to create an account.
42.
by iAkashPaul 3y ago
Expecting this to be continuously possible on a proprietary OS is a lost cause. Even the LLM world has practically mirrored this.
43.
by iAkashPaul 3y ago
TGI+grammar loaded with Mistral/Mixtral works great for structured output now! No more langchain exception handling for unmatched Pydantic definitions.
44.
by iAkashPaul 3y ago
It's basically guidelines to adhere in the output. Ex. Your prompt asks for a summary, the LLM won't necessarily just spit the summary, there might be phrases like here's the summary - then the actual summary or even basic JS
45.
by iAkashPaul 3y ago
You should do the new HF TGI server, it has both grammar & tool support now. Works fabulously with Mistral Instruct & Mixtral Instruct.
46.
by iAkashPaul 3y ago
Damn
47.
by iAkashPaul 3y ago
CC-by-NC still seems better than the DBRX's license prohibiting usage for improving other models- >2.3 Use Restrictions > You will not use DBRX or DBRX Derivatives or any Output to improve any other large language model (excludin
48.
by iAkashPaul 3y ago
I use Zoho Mail with custom domains
49.
by iAkashPaul 3y ago
His robots are just automated scripts which do the bulk repetitive tasks unlike the agents that comment OP thought.
50.
by iAkashPaul 3y ago
I'm using ReACT for RAG based chatbot & tools which are just different retrievers.
51.
by iAkashPaul 3y ago
Hey Jake, I meant capacity per shard in this case, not exhausting the IDs. Any potential solutions for that or is that not an immediate challenge?
52.
by iAkashPaul 3y ago
Even notion has a similar approach to sharding postgres but both of them could benefit from having shard IDs prefixed with YY/MM/DD(as needed) otherwise it's back to the shard navigator once they max out against org-ids for e
53.
by iAkashPaul 3y ago
Absolutely
54.
by iAkashPaul 3y ago
Reduced memory requirements, dropping higher precision IP blocks for starters
55.
by iAkashPaul 3y ago
FPGAs for native FP4 will change the entire landscape
56.
by iAkashPaul 3y ago
Seems great, coming from having to install code-server(hosted VSCode on remote machines) along side SSH via local VSCode. A better managed experience for both would be pretty neat!
57.
by iAkashPaul 3y ago
The last bit is already under way with speculative decoding but wonder what exactly are they proposing
58.
by iAkashPaul 3y ago
Despite reading it twice I couldn't come away with why they chose char/s or Hz as an appropriate measure. They also provided no benchmarks or model sizes except for a relative comparison with models 10x or 100x in size, which lead
59.
by iAkashPaul 3y ago
Cheaper phones with a better Ubuntu/Android ROMs could actually make for better SBCs when considering all sensors that come with them.
60.
by iAkashPaul 3y ago
Meta wouldn't even officially sell this let alone have repair shops in India, unlike Apple's strategy of eventually rolling out worldwide

More ›