Maple

LEARN / MODELS

Choose the model. Understand the tradeoffs.

Learn about open models and see which ones you can use with Maple today.

Explore models How Maple protects your work

Catalog reviewed

MODEL LIBRARY

Explore by family

Available in Maple Research means the model appears in the current model picker. API only models are available through Maple Proxy. Reference models are covered for learning and are not currently available through Maple.

Compare model pricing

GLM

Compare the flagship and Flash releases separately. Their capabilities and deployment requirements differ; benchmark results for one do not describe the other.

Kimi

Check the exact version and input types before choosing a Kimi model for a workflow.

DeepSeek

A separate model family with its own usage and evaluation tradeoffs.

  • DeepSeek V4.1 Flashdeepseek-v4-1-flashAvailable in Maple Research

GPT-OSS

The Safeguard variant is listed for API use only; it is not a Research model-picker option.

  • OpenAI GPT-OSS 120Bgpt-oss-120bAvailable in Maple Research
  • OpenAI GPT-OSS Safeguard 120Bgpt-oss-safeguard-120bMaple API only

Gemma

Google’s open-weight model family, with its own capabilities and deployment needs.

  • Gemma 4 31Bgemma4-31bAvailable in Maple Research

Llama

Meta’s open-weight model family for general-purpose work.

  • Llama 3.3 70Bllama3-3-70bAvailable in Maple Research

CHOOSE YOUR ACCESS PATH

The model is only part of the decision.

Where it runs affects setup, cost, data handling, and the claims you can verify.

01

Run it locally

Keep model execution on a machine you control. Check the exact weights, license, memory requirement, context setting, and performance of the version you plan to run.

See a production serving recipe
02

Use a hosted API

Let a provider run the model. Compare its pricing, retention terms, logging, and available model version before sending sensitive material.

03

Use Maple

For models we offer, use a hosted path with Maple’s client encryption and protected inference architecture. Review what the service can verify and what account or usage metadata remains.

How Maple works

DEEPER READING

Model guides

Version-specific guides, maintained as the models and Maple’s catalog change.

MODEL GUIDE

Kimi K3: Capabilities, Hardware, and Access Options

Kimi K3 is Moonshot AI's 2.8T-parameter open-weight model. See its hardware needs, how it differs from K2.6, and how to use it privately through Maple.

Read guide ↗MODEL GUIDE

GLM-5.3 Flash: What It Does and How to Use It

Learn how GLM-5.3 Flash differs from the flagship, what its FP8 recipe requires, and how local deployment compares with private access through Maple.

Read guide ↗MODEL GUIDE

GLM-5.3: Capabilities, Hardware, and Access Options

Compare GLM-5.3 with Flash, review the cited hardware recipe, and weigh local hosting against hosted access through Maple’s private inference path.

Read guide ↗