Skip to main content
Gumloop gives your agents access to the top models from every major provider. You choose the model at the top of the Agent tab.
AI models evolve rapidly. New models are usually available in Gumloop within a day of their public release, so you have the latest options even if this page has not caught up yet.

​
Choosing a model

Open the model dropdown at the top of the Agent tab. The fastest way to choose is one of the options at the top:
Agent model picker showing the Recommended and Smartest presets, a provider list, and a model detail card for Claude 4.8 Opus
  • Auto: Gumloop picks a model per message instead of pinning one. New agents normally default to Auto; Enterprise organizations and accounts where Auto is restricted default to Recommended instead. Organization default-agent settings can select another model.
  • Recommended: the best balance of speed, quality, and cost. Currently points to Grok 4.6.
  • Smartest: maximum intelligence for complex reasoning and agentic work.
Each preset maps to a current best-in-class model that Gumloop keeps up to date, so you do not have to track model releases yourself. On Enterprise plans, your organization can choose which model each preset points to, so you may see different models than the defaults. See AI Model Governance & Configuration.
On Enterprise plans, admins can restrict which models a Custom Role may use. Restricted models can appear in the picker marked Restricted by Admin, but you cannot select them. Where available, you can request access.
To pick a specific model instead, search by name or browse by provider: Anthropic, OpenAI, Google, DeepSeek, MiniMax, Z.ai, and more.
Auto suits most agents, especially when message difficulty varies. Pin Recommended or Smartest — or a specific model — when you want every run to use the same model.

​
Auto: let Gumloop choose the model

Auto is Gumloop’s model router, not a model from a provider. Select it once, and Gumloop picks which model runs each message. Your initial selection depends on your organization’s defaults and model restrictions.
Model dropdown with Auto selected, described as an advanced model router powered by Gumloop Chew

Auto sits at the top of the model dropdown on the Agent tab.

​
How Auto picks a model

1

It reads your message first

Before the agent starts working, Gumloop judges what the message actually needs — how hard it is, whether anything will check the answer, which tools it implies — and picks the cheapest tier that should still produce the same result for you. Quality comes first: Auto does not save money by accepting a worse answer. While it decides, the chat shows Choosing the best model.
2

It runs on a real model, and tells you which

Each tier is a short, ordered list of models; Auto runs the first one you’re allowed to use. Once the run starts, the chat names it (Using Grok 4.6), and that same model is what shows up in the chat’s models list and its credit breakdown.
3

It can move up mid-run

If a run turns out harder than the message looked, Auto escalates to a stronger tier part-way through — the chat shows Switching to …. That judgement is made from steps the agent has already finished, so it reacts to how a run is going rather than predicting it. Escalation then sticks for the rest of the run: a run never climbs and drops back down, and a message that already started on a top tier doesn’t switch, because nothing sits above it.
You can ask. If your own message asks for a stronger (or cheaper) model, Auto honors that instead of optimizing for cost, as long as the model you asked for can actually do the job. It only listens to your current instruction — a model name quoted inside a pasted document or a fetched page doesn’t steer the routing.

​
The tiers Auto chooses from

Auto picks one of five tiers, from cheapest to most capable. Gumloop keeps the model behind each tier current, so the models below are today’s choices rather than a fixed list — the chat always names the one that actually ran. Some work never runs on the two cheapest tiers, however trivial the message looks:
  • Anything that writes to the real world — send, post, reply, create, update, delete, invite — runs on Grok 4.6 or above, including bulk “do this to each of these” sweeps.
  • Turns that use other agents (sub-agents, delegations) run on Grok 4.6 or above, because those tools aren’t available on the cheaper tiers.
  • Messages carrying images or other non-text attachments stay on Grok 4.6, since routing only reads text and can’t judge what it can’t see.

​
Where Auto can be used

​
Auto and restricted models

On Enterprise plans, Auto stays inside your organization’s model rules — it is not a way around them.
  • Both layers of restriction apply. Organization-wide model restrictions and a Custom Role’s model allow-list are both enforced on the routing decision, on the model Auto runs, on any mid-run escalation, and on fallbacks.
  • A blocked model is never used. Auto skips it and takes the next permitted model in that tier.
  • A fully blocked tier steps aside. If nothing in the chosen tier is permitted, Auto runs on the first tier that does have a permitted model, starting with Grok 4.6’s tier.
  • If nothing is permitted, the run fails. Auto stops with a model-restricted error rather than running on a model your organization disallowed.
  • Admins can restrict Auto itself. It appears in the restriction and Custom Role model lists like any other picker entry. An allow-list that leaves it out prevents users from selecting it. Agents already saved on Auto use a permitted ordinary model, with no routing or escalation; restrictions still apply.
  • Models with 30-day data retention are never routed to. Claude Fable models have to be chosen deliberately, so Auto never selects one.

​
Auto FAQ

Auto’s router is called Gumloop Chew, and its model ID is gumloop-chew. You will see both names, so it helps to know which is which: Auto is the label in the picker, Chew is the router doing the work, and gumloop-chew is what it is called in code.
  • Set it on an agent. Set model_name="gumloop-chew" when creating or updating an agent through the API or Python SDK, or pass --model gumloop-chew in the CLI.
  • List it. GET /models returns it among the picker options, marked as a Gumloop router rather than a provider model.
  • Ask it to decide. POST /models/route returns the model Chew would pick for a message. router is always gumloop-chew; the concrete model it chose is route.model, with the lane it landed in, ordered fallback_models, and every candidate it considered.
An illustrative response is shown below. The decision depends on your message, candidate models, and permissions.
This endpoint returns a decision only. It does not run the selected model or its fallbacks. Restricted candidates may appear with status: "restricted", but are never selected or included in fallback_models.
gumloop-chew is a router, not a model you can call directly. Chat completions need a concrete model; routing is exposed through POST /models/route, and the routing judgement costs a small amount of credits. Your organization’s model restrictions apply to it just as they do in the app.
No. It’s a router. Selecting Auto means “decide for me”: Gumloop classifies each message and runs a real provider model — Claude, GPT, Gemini, Grok, and others — behind it.
The chat tells you. It shows Choosing the best model while deciding, then names the model it runs (Using …) and any mid-run switch (Switching to …). The chat’s models list and credit details record the real models, never the routing guess.
Auto won’t use it. Organization restrictions and Custom Role allow-lists apply to every model Auto could pick, including escalations and fallbacks, so a restricted model is skipped in favor of a permitted one. If no permitted model is left anywhere, the run fails with a model-restricted error instead of bypassing the rule.
Yes. Auto is listed in organization restrictions and Custom Role model lists, so it can be blocked or left out of an allow-list. Agents saved on Auto fall back to running on a normal model, without routing.
Exactly like the model that ran: token-based Chat & Reasoning, plus Compute, Tool Calls, and the orchestration fee. The routing decision itself appears as a small separate Model Routing charge in the chat’s credit breakdown.
The message runs on Auto’s baseline model — currently Grok 4.6 — instead of failing. Routing errors, timeouts, and content Auto can’t inspect all fall back to that baseline rather than guessing a cheaper or stronger model.
Use Auto for agents whose messages vary a lot in difficulty — it saves money on the easy ones without capping the hard ones. Pin a specific model or preset when you need every run to be predictable and identical, such as a compliance-sensitive agent or an eval baseline.

​
Reading the model card

Selecting or hovering a model opens a detail card so you can compare options at a glance:
  • Description: what the model is best at.
  • Speed and Intelligence: relative ratings across the catalog. Intelligence ratings are periodically rebased on a third-party benchmark, the Artificial Analysis Intelligence Index, so an existing model’s rating can change as the index is updated.
  • Provider and Context: who makes the model and how much it can read at once (in tokens and approximate words).
  • Tool Calling and Vision: capability checks.
  • Badges: extra labels that call out how a model is served. Models marked US-provider hosted are served by US providers under Zero Data Retention (ZDR) policies, and models marked 30-day data retention are not covered by ZDR.
For agents that read images or screenshots, pick one with Vision.

​
Available models

These are the models you can choose for an agent, grouped by provider. Every one supports tool calling, so it can use your apps and data. Models marked Vision can also read images and screenshots. New models are added continually, so the in-product picker is always the most up-to-date list.
All open-source models (such as GPT-OSS 120B, LLaMA, DeepSeek, Qwen, Kimi, MiniMax, and GLM) are accessed through US-based providers under Zero Data Retention (ZDR) policies, and carry a US-provider hosted badge in the model picker. Your data is never used for model training and is not stored after inference.

​
Models with 30-day data retention

Gumloop serves every model under Zero Data Retention with one documented exception: the Claude Fable family. Anthropic keeps prompts and outputs sent to Claude Fable models for 30 days to check for misuse, then deletes them. Anthropic does not train on that data, and every other model — including every other Claude model — is unaffected. These models carry a 30-day data retention badge in the picker, and:
  • They are off by default. An organization admin has to explicitly allow non-zero data retention models before anyone can select one. See AI Model Governance.
  • They are never chosen for you. They are excluded from the Recommended / Smartest presets, from Auto, and from fallback models — the only way one runs is if you pick it.
  • If the acknowledgement is later revoked, they become unavailable immediately and agents set to them fail until you switch models.

​
How agents are charged

In agents, model cost is token-based and variable. You are charged for the tokens each message uses, which depends on the model, the length of the conversation, and the tools available. There are no fixed per-message tiers. Model calls bill at cost, converted to credits at $0.005 per credit, and each run also carries Compute and an Orchestration Fee. Select the credit count in any chat’s header to see its exact usage, and see Credits for the full breakdown.

​
Bring your own key (BYOK)

Provide your own provider API key to cut model costs. BYOK makes AI model calls cost 0 credits, and it also applies to image generation and voice transcription. Since the model spend leaves your credit balance entirely, agent chats on BYOK carry a 16% orchestration fee instead of 8%, calculated on what the run would have been worth — still a large net saving. See Credits.
Each key only waives the models it can actually serve. A Fireworks AI key covers the open models Gumloop serves through Fireworks — DeepSeek V4 Flash, DeepSeek V4 Pro, Kimi K3, Kimi K2.7 Code, GLM-5.2, MiniMax M3, and Qwen3.8 Max. The other open models in the picker (GPT-OSS 120B, Gemma 4 26B, GLM-5.3-Flash) run elsewhere, so a Fireworks key does not affect them.

Setting up a key

Requires a Pro plan or higher and your own OpenAI, Anthropic, Google AI, Perplexity, SpaceXAI, or Fireworks AI account.
Add a key under your personal credentials at Connectors so your own calls route through it, or add a shared team key so the whole team can use it without managing individual keys.Pro users cannot set keys at the organization level.