[TQ-4] Create Hermes profile for local Gemma 4 + TurboQuant #28

New Issue

Timmy · 2026-04-03T22:41:27Z

Timmy commented

2026-04-03 22:41:27 +00:00

Parent: #30

Depends: TQ-1 (or TQ-2 if using llama-server directly)

Task

Create a Hermes profile that uses local Gemma 4 as its brain.

Steps

hermes profile create gemma4-local --clone
Edit config.yaml to point at local llama-server or Ollama
Write SOUL.md for the profile
Create skin
Test with a real task

Config Pattern

model:
  default: gemma4
  provider: ollama
  # OR for TurboQuant llama-server:
  # provider: custom
  # base_url: http://localhost:8081/v1

Acceptance Criteria

Profile created and doctor passes
Can chat with gemma4-local profile
Responses are coherent and useful

## Parent: #30 Depends: TQ-1 (or TQ-2 if using llama-server directly) ### Task Create a Hermes profile that uses local Gemma 4 as its brain. ### Steps 1. `hermes profile create gemma4-local --clone` 2. Edit config.yaml to point at local llama-server or Ollama 3. Write SOUL.md for the profile 4. Create skin 5. Test with a real task ### Config Pattern ```yaml model: default: gemma4 provider: ollama # OR for TurboQuant llama-server: # provider: custom # base_url: http://localhost:8081/v1 ``` ### Acceptance Criteria - [ ] Profile created and doctor passes - [ ] Can chat with gemma4-local profile - [ ] Responses are coherent and useful

Timmy self-assigned this 2026-04-03 22:41:27 +00:00

Timmy referenced this issue

2026-04-03 23:06:27 +00:00

[EPIC] TurboQuant + Gemma 4 Local Mac Deployment #30

Timmy commented

2026-04-06 04:45:27 +00:00

Stale check: this is unblocked on the TQ-1 / Ollama path now that #31 is closed, and the parent epic #30 is already closed.

I am not taking the full profile build in this bounded burn cycle. The next smallest proof-bearing slice is:

hermes profile create gemma4-local --clone
point config.yaml at the local Gemma 4 backend already validated in the TQ lane
post hermes doctor output plus one real chat transcript here

Leaving this open as the next actionable handoff instead of letting it sit silent.`

Stale check: this is unblocked on the **TQ-1 / Ollama path** now that #31 is closed, and the parent epic #30 is already closed. I am not taking the full profile build in this bounded burn cycle. The next smallest proof-bearing slice is: 1. `hermes profile create gemma4-local --clone` 2. point `config.yaml` at the local Gemma 4 backend already validated in the TQ lane 3. post `hermes doctor` output plus one real chat transcript here Leaving this open as the next actionable handoff instead of letting it sit silent.`

Sign in to join this conversation.

1 Participants

Notifications

Due Date

No due date set.

Dependencies

No dependencies set.

Reference: Timmy_Foundation/turboquant#28