DeepSeek just dropped a 5-page paper + free GitHub repo that makes any LLM respond 80% faster
it's called speculative decoding. in plain english:
Guess → Check → Keep → Repeat
> Guess: a small fast model predicts the next few words
> Check: ...
Large Language Model (LLM)
Sentiment
Shows the token's social media mindshare and sentiment over time
1D
7D
30D
90D
ALL
Our 2 main focuses on rotki right now are:
1. Refactor accounting/balance reconciliation to be easier to use and not require a PnL report
2. Analyze your rotki data with your favorite LLM via MCP
In this tough market we need help to achieve those. ...
6/26/2026 17:22
1.4K
3
5
19
0
I am really excited by what we are currently building @rotkiapp
You have this privacy preserving, rich history of your entire financial activity of the last year/decade saved in your local system
Now you can ask it questions with a local or a fron...
6/23/2026 19:31
1.5K
5
5
26
0
Someone open sourced a full LLM training pipeline implemented entirely from scratch in PyTorch — no transformers, no peft, no trl.
> Goes all the way from pretraining to post-training: Base → SFT → Reward Model → PPO/DPO → GRPO
> Trained on re...
6/21/2026 13:45
41K
20
5
89
62
GLM-5.2 is the most impressive LLM I've ever seen *aside from Fable 5*
The fact that this level of intelligence is now open-sourced is pure insanity.
Just 3 months ago, this would've been considered the most powerful coding model in the world, and ...
6/19/2026 06:34
103K
84
82
673
264
My 2 cents on @grok. People who say it's not good likely have not used it.
It's not currently the smartest LLM, but it is vastly unfucked compared to comps, it has some of the best voice features including in multiple languages, and most importantl...
6/18/2026 03:44
3.1K
9
2
33
1
🎙️ join me live tomorrow with @port_dev on slop computer
🎞️ we'll talk about his ai clipping app extractoor
🧾 but also abi ninja's Rust decoding backend (gulltoppr), vanitoor, monad-monitor TUIs, and the docusaurus "copy page for your LLM" plugi...
6/17/2026 17:32
3.8K
6
5
36
1
A LOCAL LLM RUNNING ON ONE 3090 JUST REPLACED GOOGLE HOME FOR 29 SMART DEVICES WITH ZERO CLOUD
Qwen3, Whisper and Kokoro all running at 275w. Ask if you need a jacket and it checks your actual weather sensor and answers.
https://t.co/69fRKnDZnr
6/16/2026 20:45
40K
7
4
61
12
I have a simple test I would like everyone to run.
Go to your favorite LLM and ask “how do I get my tax rate lower? Be accurate and specific.”
Then go to @cfosilvia & ask the same prompt.
Which product gives you a more valuable answer?
__
Si...
6/16/2026 15:10
22K
31
14
65
18
$RDDT - LLM-referred traffic grew sharply across Reddit, Pinterest, Facebook and Google in April–May 2026, with May an all-time high for each.
Reddit's LLM-sourced share sits around 40bps of its total traffic. Small in absolute terms but look at the...
6/15/2026 20:38
7.1K
4
3
35
6
On Wednesday June 17th I will be @dappcon 's main stage 15:30 talking about everyone's favorite topic
TAXES AND ACCOUNTING
10 yrs of crypto accounting experience w/ @rotkiapp, the necessity of data sovereignty & now with some LLM horrors sprink...
6/15/2026 11:50
1.3K
2
3
21
0
I'm seeing new trend of requests to add web sites to my Bitcoin resources list. They claim to be educational or provide various services, even claiming years of operation. But upon closer review they're LLM generated and only a few months old. 🚨
In...
6/13/2026 12:10
7.7K
6
3
67
4
We're about to run back blockspace fees.
1. Early leaders (Ethereum, Anthropic) drive BIG fees
2. Price elasticity breaks (gas, LLM budgets)
3. Cheaper 80/20 (blockspace, models) pop up
4. Revenue collapses, Jevon's paradox cited
5. Jevon's paradox ...
6/10/2026 16:20
15K
11
7
119
56
would be cool if Cloudflare could automate cost optimization via the built in observability platform - would auto generate evals / LLM judge all past threads against cheaper models vs the one used and allow you to incrementally fine tune or route to ...
6/9/2026 15:28
8.9K
11
1
43
18
GOOGLE'S GEMMA 4 12B JUST BECAME THE NEW LOCAL LLM KING FOR ANYONE WITH AN 8GB VRAM GPU
it runs at 32 tok/s with 64k context on a budget RTX card, no API, no cloud, no subscription, and it's actually smaller in file size than the previous best optio...
6/7/2026 04:45
57K
33
26
279
138
What your LLM tells you when you ask it questions about its calculations regarding your taxes (or other important matters).
"Good challenge, and you're right to push on it. Let me be straight about what I actually know here versus what I overstated"
6/3/2026 21:26
2K
4
0
10
1
Let’s say CFOs crack down on token spend, and Anthropic revenue stalls. If that happens (big if) are semiconductors screwed?
For example Micron, is there any other demand that can keep it above $1 trillion market cap outside the anticipation of exp...
5/30/2026 01:08
3.5K
14
1
21
1
SOMEONE VIBE CODED A MAC APP THAT READS YOUR FOLDER NAMES AND AUTOMATICALLY ASSIGNS THE RIGHT ICON TO EVERY SINGLE ONE.
Apple removed custom folder icons years ago and one person with an LLM just quietly brought the feature back for free.
https://t...
5/30/2026 00:15
34K
11
1
53
6
GEPA isn't just a prompt optimizer it's a way to let an LLM systematically explore a massive dataset and surface insights on its own.
The visualizer makes the process look like watching a mind work through a problem in real time.
https://t.co/5PhLs...
5/29/2026 14:15
41K
9
0
65
28
Description
#5114
Large Language Model is a crypto project available as both a token (LLM) and an NFT collection. The token works alongside a collection of 3,000 unique NFTs held by 335 owners. The project connects blockchain technology with AI language capabilities.
| Sector: | AI Agents |
| Blockchain: | Solana |
2025
Launch Year 2025
Social Statistics
Social media metrics tracking this token's online presence. Mindshare shows the percentage of crypto discussion focused on this token. Momentum indicates trend direction (rising or falling). Net Sentiment measures overall community sentiment (positive, negative, or neutral) based on the difference between positive and negative mentions.
Mindshare (24h)
0.85%
Mindshare Momentum
Down
Net Sentiment
Neutral
Sentiment Momentum
Down
Featured Tweet
Post with highest engagement and good quality mentioning Large Language Model in the past week