GPT 6 Astra and Gemini 3.8 Flash Are Live in Momen

GPT 6 Astra and Gemini 3.8 Flash are now in Momen's model list. Both are selectable from the model dropdown inside any AI Agent, with no API key and no SDK.

Both are built for long-horizon agentic work: the agent plans, calls a tool, reads the result, and keeps going. With tool calling and structured output, that means calling an Actionflow, reading a table, and handing back a typed object your app can act on — not just answering.

Understanding the Two Models

Both are text-out models, so neither generates images.

GPT 6 Astra is OpenAI's flagship, aimed at coding, research, computer use, and multi-step tasks that run for a while before they finish. Its 128,000-token output ceiling is the one to note: it matters when a single step has to emit a long document or a large structured payload in one call.

Gemini-3.8-flash is Google's current Flash model, built for agent and software-engineering work at Flash speed and cost.

Two things to know before wiring Gemini-3.8-flash in. It accepts audio and PDF at Google, but a Momen agent takes text, image and video inputs, so those two are out of reach. And its thinking levels are tunable through Google's API but not exposed in a Momen agent, so it runs at the default.

How to Use Them in Momen

Both are wired in the same way, as the model behind an AI Agent.

1. Open the AI page in the left toolbar and go to the Agent tab.

2. Click Add agent, or open an agent you already have.

3. Click the model name at the top of the agent editor and choose GPT 6 Astra or Gemini-3.8-flash. The same panel holds Temperature, Max number of rounds, Maximum output tokens, and Image processing.

4. Write the System and User halves of the Prompt template. Add Inputs for anything the prompt needs as a variable, and attach Contexts — Data or API — for knowledge the agent should retrieve at run time.

5. Under Tools, give the agent what it is allowed to act on: an Actionflow, an API, or another Agent.

6. In Outputs, choose Plain text or Structured, and switch on Streaming output if your UI should stream tokens.

7. Test it under Debug & preview with Start conversation.

The agent then drops into an Actionflow like any other node, so AI output can trigger a database write, an email, or a payment. The AI integration guide covers that wiring, and the Actionflow node list covers what it can chain into.

What It Costs

Both models are billed in AI Points.

Point conversion

Gemini-3.8-flash and Gemini-3.7-flash carry identical rates, so switching an existing agent between the two does not change what it costs.

What one agent call costs

The formula is (input tokens × input conversion) + (output tokens × output conversion).

Take a realistic agent call: a 2,000-token input — system prompt, the user's message, and a few rows of retrieved context — and a 500-token answer. On GPT 6 Astra:

Input: 2,000 × 8.5700 = 17,140 points Output: 500 × 42.8600 = 21,430 points Total ≈ 38,570 points

The same call across the list:

Astra costs about 1.8 times a GPT 5.6 Sol call (38,570 ÷ 21,435), which makes it the most expensive option in Momen's list. Gemini-3.8-flash costs about 3.4 times a GPT 5.6 Luna call (2,885 ÷ 855), leaving it the second-cheapest row in the table.

How far the Free plan goes

The Free plan includes 100,000 AI Points per month. Against the call above:

GPT 6 Astra: 100,000 ÷ 38,570 ≈ 2 calls GPT 5.6 Sol: 100,000 ÷ 21,435 ≈ 4 calls Gemini-3.8-flash: 100,000 ÷ 2,885 ≈ 34 calls GPT 5.6 Luna: 100,000 ÷ 855 ≈ 116 calls

Two calls is enough to see Astra answer and not much else. Points top up at 6,000,000 for $10, which puts one point at $10 ÷ 6,000,000. A 38,570-point Astra call is about $0.064; the same call on Gemini-3.8-flash is about $0.005.

Plans and top-ups

Basic includes 1,000,000 AI Points per month and Pro includes 5,000,000. On the same 2,000-in / 500-out call, Basic covers roughly 25 Astra calls or 346 on Gemini-3.8-flash; Pro covers roughly 129 and 1,733. Plan details are on the pricing page, and the usage calculator will size a specific workload with your own token counts.

Real Showcases

These are apps already running on Momen where the agent, not a chat box, is the working part.

A multi-round AI judge. The Momen team built an agent that runs several rounds of dialogue before returning a verdict, each round feeding the next. It is the clearest example of an agent as a multi-step process rather than a single call. Read more

Resume-to-job match scoring. Rajeevdaz built an app that scores a resume against a specific job posting before the user applies. The agent has to return a consistent, comparable score rather than an opinion, which is exactly what structured output is for. Read more

Photo-based junk removal quotes. BayHaul prices a haul from photos a customer uploads instead of a phone call. The agent reads the images and the description together and returns a price range with the items it detected. Read more

Conclusion

A new model in the list is a smaller event than it looks, because the model was never the hard part. What turns an AI feature into a product is everything wrapped around the call — a database that keeps the answer, an Actionflow that decides what happens next, accounts so the right person sees it, storage for the files, payments if you charge for it — and that is what Momen gives you already built. Start from the app you want to ship, and treat the model as the setting you can change later.