anomalyco / anomalyco/opencode

Why is Nemotron 3 Ultra the only model that keeps its 1M context on the free tier?

Open
#40,928 1 comment 0 reactions 1 assignee View on GitHub

@fwang is already working on this.

Since Aug 6, 2026.

Dominant language
TypeScript
Stars
209k
Forks
27.5k
PR merge metrics
PR metrics pending

Description

Description

Multiple models on the Zen free tier natively support 1M context but get capped at 200-300K. DeepSeek V4 Flash (native 1M) is stuck at ~200K. Laguna S 2.1 (native 1M) is stuck at ~270K. But Nemotron 3 Ultra easily the worst model on the platform somehow keeps its full 1M context on the same free tier.
This was previously reported in #27929 when DeepSeek V4 Flash lost its 1M context, and nothing changed.
It would be nice if either:

  1. All models got their native context windows back on the free tier, or
  2. If Nemotron has a special deal or different terms, that should be documented somewhere. Right now it just looks arbitrary.
  3. Also the Zen models page does not show capped vs native context lengths, so you have to discover the hard way that your 1M model is actually 200K.
Plugins

none

OpenCode version

1.18.10

Steps to reproduce

No response

Screenshot and/or share link

No response

Operating System

No response

Terminal

No response

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.