[Upstream PR #92] Add DeepSeek context fallback

[quangdang46]

Mirrored from upstream 1jehuang/jcode — Pull Request #92 by @MukundaKatta
Original state: open
Created: 2026-04-30T21:05:39Z · Updated: 2026-04-30T21:05:43Z
Diff: https://github.com/1jehuang/jcode/pull/92.diff
This issue is an auto-mirrored copy. Comments and edits here are local to quangdang46/jcode — do not expect them to propagate upstream.

---

Summary

• add a hardcoded 1M token fallback for DeepSeek model IDs
• cover deepseek-v4-flash and deepseek-chat in context limit tests

Fixes #87.

Tests

• cargo test test_context_limit_deepseek

---

^{Need help on this PR? Tag @codesmith with what you need.}

• Let Codesmith autofix CI failures and bot reviews

[quangdang46]

Covered by PR #181 which adds the same model.starts_with("deepseek") -> Some(1_000_000) branch (in crates/jcode-provider-core/src/models.rs since this fork moved the fallback_context_limit_for_model body into the core crate) plus a regression test context_limit_falls_back_to_1m_for_deepseek_models. Upstream PR #92 (this issue) and upstream PR #102 (issue #87) both proposed the same DeepSeek branch; PR #181 ports it once.

Closing as superseded by #181.