grok-4.6 context window is reported as 200k instead of 500k
Nobody has claimed this yet.
- Dominant language
- Rust
- Stars
- 19.9k
- Forks
- 2.3k
- Avg merge
- 2d 7h
- Merged PRs (30d)
- 30
Description
Summary
xAI documents Grok 4.6 / 4.5 as a 500,000-token context window. Jcode has no family entry for those ids, so context_limit_for_model("grok-4.6") falls through to DEFAULT_CONTEXT_LIMIT (200,000). The TUI meter and compaction budget then under-report the real window.
Expected
grok-4.6 and grok-4.5 resolve to 500k. grok-code-fast stays 256k.
Actual
200k default.
Notes
This is a client-side catalog/heuristic issue only. The model still cannot take 1M; 500k is the official cap.
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start at context_limit_for_model("grok-4.6") and trace the model catalog or heuristic used when resolving model IDs. Check how the TUI meter and compaction budget consume that limit. Done means grok-4.6 and grok-4.5 resolve to 500,000 tokens while grok-code-fast remains at 256,000.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- rust
- Domain
- ai, cli
- Issue type
- Bug
- Difficulty
- 2/5
- Estimated time
- 1-3 hours
- Activity status
- Active
- Clarity
- Clearly specified
- Newbie friendliness
- 78/100