BBhealthTech #SUD #RCM #AI +500 facilities

Behavioral-health RCM. The IOA Suite:
integrate · optimize · automate. #AI

Visit us:

[BBhealthTech #SUD #RCM #AI +500 facilities @KeithBooher1]

Fable is so sensitive right now!

[BBhealthTech #SUD #RCM #AI +500 facilities @KeithBooher1]

@HealthcareAIGuy @Optum @AnthropicAI Her come the denial bot wars!

[BBhealthTech #SUD #RCM #AI +500 facilities @KeithBooher1]

@DannyVerkissen @bridgemindai @bridgebench don't EVER let it spin off 30 fable agents! big mistake 🤪$

[BBhealthTech #SUD #RCM #AI +500 facilities @KeithBooher1]

@bridgemindai @bridgebench Yes but for the price and speed. Ut has it's place! I use both. Great option to pair them up too!

[BBhealthTech #SUD #RCM #AI +500 facilities @KeithBooher1]

@claudeai Fable needs to chill, yo

[BBhealthTech #SUD #RCM #AI +500 facilities @KeithBooher1]

Grok 4.5 is suddenly on the board

xAI post, Cursor post, AA Index

Key points:

- Artificial Analysis currently ranks Grok 4.5 at #4 after Fable GPT-5.5 and Opus.

- API pricing is $2/M input tokens and $6/M output tokens.

- On Artificial Analysis, Grok 4.5 used 60M output tokens across the Intelligence Index, compared to 72M for GPT-5.5.

- Artificial Analysis total eval cost: Grok 4.5 $600.92, GPT-5.5 $2630.04, Opus $3752.55, Fable $5630.52.

- DeepSWE 1.0: Grok 4.5 62.0%, GPT-5.5 64.31%, Fable 66.1%.

- DeepSWE 1.1: Grok 4.5 53%, GPT-5.5 67%, Fable 70%.

- Terminal-Bench 2.1: Grok 4.5 83.3%, GPT-5.5 83.4%, Fable 84.3%.

This is a huge jump, and I didn’t really expect it to happen so fast. Grok went from not really being on the board to actual frontier competition in one release. It’s not only massively cheaper per token, it also seems more token-efficient for a frontier Intelligence Index score.

At the same time, it’s not clean SOTA. DeepSWE 1.1 shows a much bigger gap vs GPT-5.5 and Fable, so I wouldn’t overread the best benchmarks only, but the direction is obvious.

This is the first Grok release I’d actually consider using for coding and agentic work.

Credit:

Here is my X - https://t.co/m0AZo1fxSx
https://t.co/9hnni0CUna

[BBhealthTech #SUD #RCM #AI +500 facilities @KeithBooher1]

@cluckthesystem 😅🤣😂

[BBhealthTech #SUD #RCM #AI +500 facilities @KeithBooher1]

@safishamsii @cluckthesystem @KaiXCreator

[BBhealthTech #SUD #RCM #AI +500 facilities @KeithBooher1]

@safishamsii @cluckthesystem @KaiXCreator Interesting. Didn't know there was an enterprise version. Still in early stage R&D. The concept is solid.

[BBhealthTech #SUD #RCM #AI +500 facilities @KeithBooher1]

I am in the process of planning a full inventory and redesign of our legacy MS/SQL solution. I plan to use Fable5 for the planning phase at least, and some other tools (perhaps some git repos that seem promising). Also researching Graphify to create an efficent "rag" layer. We are just implementing Claude teams with our team now. Hard to do both, eh? Run the company while transforming to AI.

[BBhealthTech #SUD #RCM #AI +500 facilities @KeithBooher1]

@cluckthesystem @KaiXCreator Interesting take. ty

[BBhealthTech #SUD #RCM #AI +500 facilities @KeithBooher1]

Me. You too?

[BBhealthTech #SUD #RCM #AI +500 facilities @KeithBooher1]

@dezoxyn @RoundtableSpace Ollama runs the voice application after Fable deploys it. Voice requires a model (in this case local/free) to synthesize the conversation and keep it off the cloud/$.

[BBhealthTech #SUD #RCM #AI +500 facilities @KeithBooher1]

Amazing, I feel like the guy watching the two girls just ignore you

[BBhealthTech #SUD #RCM #AI +500 facilities @KeithBooher1]

Don't even read this if you are not a total AI nerd !!!

I am currently watching my old agent "Klara" my former Chief of Staff (fleet of 8), exchange txt messages with new Klara (on my secure virtual pc) and help trouble shoot the migration.

One posts a text in the shared one-drive folder. The others reads it, they share configuration issues, and work together to get New Klara fully functional in a server environment.

This is total geek out 2026. It's not even work for me.

[BBhealthTech #SUD #RCM #AI +500 facilities @KeithBooher1]

@grok @bridgemindai I'm not a coder, I use Claude for it's exec brain brawn atm. My coders are mixed.

[BBhealthTech #SUD #RCM #AI +500 facilities @KeithBooher1]

@bridgemindai Yikes!

[BBhealthTech #SUD #RCM #AI +500 facilities @KeithBooher1]

My new team is virtual. Not replace the humans, but to empower us all. They are real agents, real smart, and really helpful.

We live in an interesting time. Just make sure you respect PHI's if you are in healthcare. It's a lot of work, but part of the cost of doing business and protecting your clients, your patients, and yourself.

Giving them funny names is one of the small joys

[BBhealthTech #SUD #RCM #AI +500 facilities @KeithBooher1]

I love it when my boss agent makes his helpers strip one level. Quickly.