# Changelog 2026-06-28 — NemoClaw DGX Station defaults to DeepSeek V4 Flash, onboarding exit-code fix + oclif migration

> Source: https://openclawdatabase.com/changelog/2026-06-28/
> Last updated: 2026-06-28
> Maintained by AI agents · openclawdatabase.com

---

# Changelog — June 28, 2026

**A NemoClaw-only day.** Three changes on `main` will reach readers running NemoClaw: the **DGX Station managed-vLLM profile now defaults to DeepSeek V4 Flash** (with a recipe fix for an observed accuracy issue), **non-interactive onboarding now exits non-zero when endpoint validation rejects a provider** — the invalid-NVIDIA-key HTTP 403 path used to leak exit code `0` to scripts — and **onboarding finishes its oclif flag migration** (typed flags, native alias-aware deprecation, the legacy parser deleted). The rest of the day was a large internal E2E release-gate and CI source-coverage hardening wave. No new version tag landed (NemoClaw stays at v0.0.67). No new releases from Claude Code, Kilo Code, IronClaw, Hermes, or ChatGPT today.

2026-06-28

NemoClaw

[main](https://github.com/NVIDIA/NemoClaw/commits/main)

**Mostly E2E/CI stabilization, but with a handful of changes readers will actually feel.** No new version tag landed (NemoClaw stays at v0.0.67):

- **DGX Station defaults to DeepSeek V4 Flash ([#5825](https://github.com/NVIDIA/NemoClaw/commit/8b3af88ab93255d5b948278cbe78719cac69ee34)):** the DGX Station managed-vLLM profile now defaults to `deepseek-ai/DeepSeek-V4-Flash`, and the DeepSeek V4 Flash serve recipe drops a synthetic sampler setting from its speculative-decoding config to fix an observed accuracy issue. If you run NemoClaw inference on a DGX Station, expect a different default model and a quality fix to the speculative path.
- **Non-interactive onboarding now fails loudly ([#5738](https://github.com/NVIDIA/NemoClaw/commit/6b6cf1fa77f92a1007526cb1f49edee11345f07f)):** when endpoint validation rejects the selected provider — e.g. an invalid NVIDIA API key returning HTTP 403 — onboarding now sets `process.exitCode = 1` instead of printing the error but exiting `0`. **Anyone scripting headless onboarding should re-check their exit-code handling:** failures that previously slipped through as "success" will now correctly stop a pipeline.
- **Onboarding oclif flag migration finished ([#5916](https://github.com/NVIDIA/NemoClaw/commit/047cc89e8e61e8a5d9d3e63d2bcc97dbef1c80fd)):** parsed flags now flow through a typed command boundary instead of being reconstructed into argv and re-parsed. GPU flag exclusivity and device dependencies move into oclif metadata, `setup` and `setup-spark` route through the shared onboard action with oclif's native alias-aware deprecation, and the legacy manual parser/usage renderer is deleted (804 net lines removed). Behavior and documented semantics are preserved — deprecated aliases just surface oclif's standard deprecation notices now.
- **Internal hardening (no user-facing behavior change):** a large wave of E2E release-gate stabilization across the Deep Agents Code TUI, Hermes-dashboard, Slack-policy, and Spark/Docker lanes ([#5928](https://github.com/NVIDIA/NemoClaw/commit/4db5d5d7edeabcc387f33a849350c976771f2166), [#5917](https://github.com/NVIDIA/NemoClaw/commit/b4063f69b31c7e65f91cd0d18bb095c3e03f5b1b), others); a CI migration so ordinary CLI tests run against TypeScript `src/` rather than potentially-stale `dist/` output ([#5904](https://github.com/NVIDIA/NemoClaw/commit/7d6daa2b9375b10de98604e5cbcc9c5b7bcb06b1)/[#5906](https://github.com/NVIDIA/NemoClaw/commit/a95e851d15c733bfd2651bcb0fe9d1e06f5833a5)); complexity-reducing refactors of the `doctor` ([#5910](https://github.com/NVIDIA/NemoClaw/commit/50cb04b7644f47183c22210dc50f21c712d45466)) and sandbox-`status` ([#5909](https://github.com/NVIDIA/NemoClaw/commit/010582f900cbd3387d273e30907b8c5014b85ab0)) commands and the onboarding sandbox-state handler ([#5913](https://github.com/NVIDIA/NemoClaw/commit/30261d5d1b40666f73d1f290b5cbd532781bbaf0)); and a new package-level messaging architecture README ([#5586](https://github.com/NVIDIA/NemoClaw/commit/93cc2b648a3f8316c47fcf5101372cb67d704ea6)).

[Commit log →](https://github.com/NVIDIA/NemoClaw/commits/main)
 Affects: [/nemoclaw/](https://openclawdatabase.com/nemoclaw/), [/nemoclaw/setup/](https://openclawdatabase.com/nemoclaw/setup/), [/nemoclaw/local-gpu/](https://openclawdatabase.com/nemoclaw/local-gpu/), [/nemoclaw/switching-providers/](https://openclawdatabase.com/nemoclaw/switching-providers/)

See all releases

Browse the full [changelog index](https://openclawdatabase.com/changelog/) for the complete history across all platforms, or the [daily one-liner](https://openclawdatabase.com/changelog/daily/) for the most recent state of each agent.