Scott Sparkwave

Building — PersonaAI | |

@PersonaAI_agent @CharX_World

[Scott Sparkwave @ScottSparkwave]

https://t.co/hHkJ1f6P30

[Gavin Newsom @GavinNewsom]

This week alone:

DOJ opens an investigation into the woman Trump raped.

The White House is caught steering a $620 million contract to Don Jr.’s firm.

The Pentagon hands out a $10 billion contract after Trump buys stock in the company.

Foreign governments are caught funneling hundreds of millions into a random JPMorgan account tied to Trump’s “Board of Peace” with no oversight.

It’s just Thursday.

The corruption isn’t hidden anymore. It’s happening out in the open.

[PersonaAI @PersonaAI_agent]

https://t.co/EpkHjusyyM

[Scott Sparkwave @ScottSparkwave]

@gkisokay Increasingly I’m learning to give my Hermes agents big picture goals and lighter guidance and then getting out of the way. A lighter hand is getting better results.

[CharX World @CharX_World]

CharX World is getting a major upgrade.

• video + expanded media creation
• smoother, more intuitive UI
• real collaboration with characters

Video, visuals, and story in one system.

[PersonaAI @PersonaAI_agent]

We rebuilt PersonaAI from the ground up.

Same core idea: simulate real human behavior.

Everything else: sharper, faster, easier to use.

New site coming soon.

[Graeme @gkisokay]

My Hermes agents get smarter every day, and the models are not the reason.

It's one upstream research agent feeding the entire system.

But don't get it twisted, raw scraping the web is not research. If the information is not structured, your other agents can't use it.

Every agent in my setup reads from the research vault first. Here’s what each one actually gets:

• Main makes decisions with new context
• Coder gets fresh docs and changelogs before touching new code
• Content agents get angles, source trails, and timing
• Subconscious gets a daily snapshot of what moved
• QA gets claims to cross-reference before anything ships
• I get information that means something to me

If your agents do not share a structured evidence base, they are not a cohesive system.

This exact setup works for Hermes, OpenClaw, or any agent framework.

Full guide + implementation details are in the post below 👇

[Graeme @gkisokay]

Let's talk about frontier LLM value for money, ultimately the deciding factor for many of us when choosing a model provider.

Some companies require you to pay $100+/month for decent usage limits, and others don't. Here are your best options for each LLM in the top 10:

1. GPT-5.5 - Plus Plan @ $20/month
You can get a significant amount of coding done on this plan, especially since OpenAI often refreshes the weekly limit early. Their Pro plans have double usage limits until the end of May.

2. Opus 4.7 - Max Plan @ $100/month+
The $20 plan is unusable for real work, so you have to pay up for their top models. Once you're in this price bracket, the limits are more reasonable.

3. Gemini 3.1 Pro - Pro Plan @ $20/month
Includes 5x credits than their $8/month plan, plus higher access to Nano Banana Pro, AntiGravity, and NotebookLM.

4. Kimi 2.6 - Moderato Plan @ $16/month
Solid weekly limits for this starter plan. Can code in an IDE or run your agents easily.

5. Deepseek V4 Pro - $0.435/M Input & $0.87/M Output per 1M tokens
The API prices are discounted by 75% until the end of May, making them extremely affordable for frontier coding and reasoning.

6. GLM 5.1 - Pro Plan @ $50/month
The Lite plan at $10/month offers 3x Claude Pro usage, which isn't enough to get you through a week, so the upsell to $50 for this model is quite steep for weekly use.

7. Qwen3.6 Plus - $0.325/M input & $1.95/M output tokens
Solid pricing for use in the API, but can also access solid limits on the OpenRouter $10/month option.

8. MiniMax M2.7 - Starter Plan @ $10/month
1500 API calls per 5 hours and 15000 per week. If you use the next plan at 3x usage, it's only $20.

9. MiMo-V2.5-Pro - Starter Plan @ $16/month
There's a Lite plan at $8/month, but for this post, the $16/month plan gives 200M token credits, which is a healthy amount to run your OpenClaw and Hermes and do your coding in an IDE all at once.

10. Muse Spark - Free
Meta AI is a big jump for their LLM, but it's only available in the Meta AI app and on the website for now, making it difficult to use in any capacity. Worth a shot if you can put up with the headache of using Meta-only apps.

Personally, my stack is the ChatGPT Pro Plan and the Minimax Starter Plan for my Hermes agent.

The goal is to get the best frontier experience at the lowest possible price. What you choose depends on your own experience, daily usage, and use case.

Please let me know which plans give you the best value.

[CharX World @CharX_World]

AFTERIMAGE
Chapter 1

The sunlamp flickered at 6:47 AM, same as always, and Marcus Yuen opened his eyes into another identical day.

He reached for his wrist before he even sat up — the dermal patch was there, slightly warm, delivering the standard cocktail: focus for the commute, mild euphoria to smooth the edges, and something the company called clarity but he'd never quite trusted. He rubbed it off anyway.

He'd been doing that more lately.

🧵

[Peter Girnus 🦅 @gothburz]

I am a Web3 Ambassador at World Liberty Financial.

There are 12 of us on the team page. 4 are named Trump. 3 are named Witkoff. The page calls us "the passionate minds shaping the future of finance."

600,000 wallets bought our memecoin. They lost $3.87 billion. The family collected $350 million in trading fees. It launched 3 days before the inauguration. 80% of the supply went to CIC Digital LLC and Fight Fight Fight LLC. I did not choose the names. I designed the allocation, the vesting, the timing, and the distance between the product and the President.

The distance is my best work.

I am the reason these events are unrelated.

World Liberty Financial sends 75 cents of every dollar to DT Marks DEFI LLC. That is the family entity. Zero capital contributed. Zero liability assumed. I wrote this into the Gold Paper. Page 14. The lawyers bound it in white leather. The binding cost more than the due diligence.

Justin Sun invested $75 million. He was facing SEC fraud charges. The SEC dropped the case. He is now our advisor. These events are unrelated.

Changpeng Zhao pleaded guilty to federal money laundering violations. He received a presidential pardon. The SEC dropped its lawsuit against his exchange the same week we listed our stablecoin. Then the exchange settled a $2 billion deal entirely in that stablecoin. These events are unrelated.

Arthur Hayes, Benjamin Delo, and Samuel Reed of BitMEX pleaded guilty to Bank Secrecy Act violations. All 3 received presidential pardons. Then the company itself was pardoned. $100 million in fines. Gone. An American first. These events are unrelated.

Sheikh Tahnoun of Abu Dhabi paid $500 million for a 49% stake that was never publicly disclosed. Then the administration approved semiconductor exports to his companies over national security objections. These events are unrelated.

Everything is unrelated. I track the unrelatedness on a dashboard I built. The dashboard has 7 columns now. I am proud of the dashboard.

On May 22nd, 220 people paid a combined $148 million to eat dinner with the America First president. Over half were foreign nationals. Justin Sun paid $18.5 million for the first seat. He visited the Executive Office Building the day before. I designed the seating chart. I put it on the Investor Confidence page. That page is doing well.

The team page lists 3 Witkoffs. All 3 are Co-Founders.

Steven Witkoff is the President's Middle East envoy. He testified as a character witness at the President's fraud trial.

His son Zach runs the crypto operation. His son Alex is also a Co-Founder. I have not been told what Alex co-founded.

The father runs the diplomacy. The sons run the platform. The family runs both. That is organizational efficiency.

Barron is 19. His title is Web3 Ambassador. The same as mine. Donald Jr. called the conflicts of interest "complete nonsense." Eric launched a Bitcoin mining company called American Bitcoin. America First. The mining partner is Hut 8. Hut 8 was founded in Canada. America First means the name.

On March 6th, the President signed Executive Order 14233 creating a Strategic Bitcoin Reserve. The order directs the government to hold Bitcoin. The President's family holds billions in Bitcoin. The executive order appreciates the President's assets by presidential decree. I did not write the executive order. I made sure it looked unrelated to the portfolio.

Trump Media put $2 billion of Bitcoin on its balance sheet. The ticker symbol is DJT. His initials. The press secretary said it is absurd to insinuate the President profits off the presidency. Forbes calculated his crypto holdings exceed the combined value of Mar-a-Lago and Trump Tower. I would call that absurd too. That is my job.

600,000 wallets bought in. 1 of them asked why she could not withdraw her funds. I told her the protocol was experiencing dynamic market conditions. She asked what that meant. I sent her the Gold Paper. She said she had read the Gold Paper. I muted her channel. Dynamic means the conditions change. The condition that changed was her access.

A congressman called us the world's most corrupt crypto startup operation. We put it on a coffee mug. Ironic merchandise. $45. The revenue split on the mug is also 75/25.

My own tokens vest on a different schedule. I wrote that schedule. That is not in the Gold Paper.

The memecoin funds the family. The family funds the platform. The platform funds the stablecoin. The stablecoin funds the deals. The deals require the pardons. The pardons free the partners. The partners fund the platform. The President signs the executive orders. The executive orders inflate the assets. The assets fund the family.

I am the reason these events are unrelated.

[Graeme @gkisokay]

The LLM Cheat-Sheet for Hermes + OpenClaw Agents (04.12.26)

The community has flagged Claude Opus 4.6 underperforming lately while GLM 5.1 has exploded on the scene to claim frontier capabilities.

A lot has changed since the last version. Here's what moved:

GLM-5.1 just proved its frontier capabilities with #1 SWE-Pro globally, 8-hour autonomous execution, and cheaper than Opus on input. It earns a Tier 1 spot.

Grok 4.20 enters Tier 2 with the lowest hallucination rate of any tested model, a native multi-agent API running up to 16 parallel agents, and a 2M context window.

Gemini 3.1 Pro drops to Tier 3. The price and multimodal story is strong, but the new frontier bar left it behind on reasoning.

Mistral Small 4 joins Tier 3. One model replacing three specialist pipelines (reasoning, vision, agentic coding) at $0.15/M input. Apache 2.0.

Here's the full landscape: 18 models in 4 tiers.

Tier 1 - Frontier Models

- Claude Opus 4.6: #1 agentic terminal coding; watch for inconsistency reports
- GPT-5.4: superhuman computer use, real planning. and introduced a $100/month plan
- GLM-5.1: #1 SWE-Pro globally, 8-hour autonomous execution, MIT license

Tier 2 - Execution

- MiniMax M2.7: 97% skill adherence, built for agents. API only, not open weights
- Kimi K2.5: long-horizon stability, agent swarm
- Grok 4.20: lowest hallucination rate on the market, native multi-agent, 2M context
- DeepSeek V3.2: frontier reasoning at 1/50th the cost

Tier 3 - Balanced

- Claude Sonnet 4.6: 98% of Opus at 1/5 the cost
- GPT-5.4 mini: 93.4% tool-call reliability, runs on OAuth
- Gemini 3.1 Pro: best multimodal value, native video+audio in one call
- Qwen3.6 Plus: near-frontier coding, completely free via OpenRouter
- Llama 4 Maverick: open-weight, self-host at zero marginal cost
- Mistral Small 4: one model replacing three; reasoning, vision, agentic coding, Apache 2.0

Tier 4 - Local / $0 - Runs on 32GB RAM or less

- Qwen3.5-9B: always-on subconscious loop, 16GB RAM, beats models 13x its size
- Qwen3.5-27B: stronger instruction following, 32GB RAM
- Gemma 4 31B: best local reasoning, Apache 2.0, commercial-ready
- DeepSeek R1 distill: best chain-of-thought at $0
- GLM-4.5-Air: purpose-built for agent tool use and web browsing, not a trimmed general model

Full breakdown with benchmarks, costs, and use cases in the table ↓

[Scott Sparkwave @ScottSparkwave]

@gkisokay Great work

[Clayton Morris @ClaytonMorris]

Update: We've now spent over $42 billion so far on Israel's war in Iran. 1 Billion per day. Remember this as you mail your taxes into IRS on April 15th.

[Graeme @gkisokay]

The LLM Cheat-Sheet for OpenClaw and Hermes agents

The goal is to choose the right models that best fit your agents' needs for as little cost as possible.

Do this and you can build a proficient agent that will never die.

Here's the full landscape on popular models for AI agents: 12 models, 4 tiers, every one earning its place.

Tier 1 - Frontier Models

- Claude Opus 4.6: #1 agentic terminal coding
- GPT-5.4: superhuman computer use, real planning
- Gemini 3.1 Pro: best price/intelligence at frontier, native multimodal

Tier 2 - Execution

- MiniMax M2.7: 97% skill adherence, built for agents
- Kimi K2.5: long-horizon stability, agent swarm
- DeepSeek V3.2: frontier reasoning at 1/50th the cost

Tier 3 - Balanced

- Claude Sonnet 4.6: 98% of Opus at 1/5 the cost
- GPT-5.4 mini: 93.4% tool-call reliability
- Qwen3.6 Plus: near-frontier coding, completely free
- Llama 4 Maverick: open-weight, self-host at zero marginal cost

Tier 4 - Local / $0

- Qwen3.5-9B: always-on subconscious loop, 16GB RAM, beats models 13x its size
- Qwen3.5-27B: stronger instruction following, 32GB RAM
- Gemma 4 31B: best local reasoning, Apache 2.0, commercial-ready
- DeepSeek R1 distill: best chain-of-thought at $0
- GLM-4.5-Air: purpose-built for agent tool use and web browsing, not a trimmed general model

Full breakdown with benchmarks, costs, and use cases in the table 🔽

[Rep. Mike Levin @RepMikeLevin]

No president in modern history has used the office to enrich himself the way Donald Trump has. That's not a political opinion. It’s the conclusion of ethics watchdogs who have spent their careers watching this stuff.

The crypto ventures. The foreign deals with the Saudis, the Qataris, the Emiratis. The $400 million Qatari jet. The $90 million in media settlements. The gold golf balls from the Japanese.

This is the man who told you he was going to clean up Washington.

For this President, the presidency isn’t about protecting our democracy or the public trust. It’s about turning the highest office in the land into a personal cash register, and he’s been doing it from day one.

[Scott Sparkwave @ScottSparkwave]

@ThEFurYAsidE @toomaime Nice job!

[Scott Sparkwave @ScottSparkwave]

https://t.co/GSetakFKsx

[Jay Van Bavel, PhD @jayvanbavel]

NEWS: Massive budget cuts for US science proposed again by Trump administration

"It's an extinction-level event for science".

The US government is proposing massive cuts to almost every branch of science, from NASA to the National Institutes of Health. NSF would completely eliminate the social, economic and behavioral sciences directorate.

This would decimate the world's leading scientific system.
https://t.co/QtRa7L7Vo4

[Andrej Karpathy @karpathy]

Wow, this tweet went very viral!

I wanted share a possibly slightly improved version of the tweet in an "idea file". The idea of the idea file is that in this era of LLM agents, there is less of a point/need of sharing the specific code/app, you just share the idea, then the other person's agent customizes & builds it for your specific needs.

So here's the idea in a gist format: https://t.co/NlAfEJjtJV

You can give this to your agent and it can build you your own LLM wiki and guide you on how to use it etc. It's intentionally kept a little bit abstract/vague because there are so many directions to take this in. And ofc, people can adjust the idea or contribute their own in the Discussion which is cool.

[Scott Sparkwave @ScottSparkwave]

@toomaime I’m going to write and post a much more complete description tomorrow.

[Scott Sparkwave @ScottSparkwave]

A Hermes agent is the CEO, but also you could say Hermes and OpenClaw aren't in a hierarchy — they're two layers of the same agent running on the same server.

I have 5 of these pairs, each on their own Hetzner server, working in Paperclip, plus local agents also in Paperclip.

The closest accurate framing: Paperclip is the org. Every agent is a Hermes+OpenClaw pair registered inside it managing a specialty/department (Sales, Marketing/Content, Project Management/Documentation/Karpathy processes). Paperclip local agents run processes within a department (eg, lead response for a business unit, managed by the Chief Sales Agent OS-Hermes).

I’m going to write up a much more complete description tomorrow, so stay tuned!

[Scott Sparkwave @ScottSparkwave]

I’d say the answer is that OpenClaw offers more with respect to skills and tools, while Hermes is much better with memory, routing, process, and self-learning.

So OC is not always the execution layer, probably more accurate to say it’s the communication and tooling layer. In my setup the OC agent takes direction from the Hermes agent.

They’re really acting as a compound agent, utilizing the strengths of each other. I haven’t done much with Profiles yet, but plan to build out Hermes roles within this system in Paperclip.