Skip to content

Fix Fireworks "AMD-hosted" Contradictions#2

Open
Nevern1y wants to merge 2 commits into
mainfrom
fix-amd-hosted-claims-6386774088937261027
Open

Fix Fireworks "AMD-hosted" Contradictions#2
Nevern1y wants to merge 2 commits into
mainfrom
fix-amd-hosted-claims-6386774088937261027

Conversation

@Nevern1y

@Nevern1y Nevern1y commented Jul 8, 2026

Copy link
Copy Markdown
Owner

Removed inaccurate claims that the Fireworks backend is AMD-hosted. Clarified that the true AMD proof relies on the self-hosted MI300X or W7900 setups.


PR created automatically by Jules for task 6386774088937261027 started by @Nevern1y

Identified a contradiction where Fireworks was labeled as "AMD-hosted"
in multiple places (`README.md`, `AGENTS.md`, `app.py`, etc.), while
`DEPLOY.md` and `docs/real_runs/README.md` correctly state it runs on
NVIDIA H100 GPUs.

To comply with the "Honest copy: no fake AMD claims" design rule, removed
"AMD-hosted" descriptions for the Fireworks backend across the codebase.
Clarified that only the self-hosted MI300X/W7900 setups count as true
AMD compute proof.

Co-authored-by: Nevern1y <77607229+Nevern1y@users.noreply.github.com>
@google-labs-jules

Copy link
Copy Markdown
Contributor

👋 Jules, reporting for duty! I'm here to lend a hand with this pull request.

When you start a review, I'll add a 👀 emoji to each comment to let you know I've read it. I'll focus on feedback directed at me and will do my best to stay out of conversations between you and other bots or reviewers to keep the noise down.

I'll push a commit with your requested changes shortly after. Please note there might be a delay between these steps, but rest assured I'm on the job!

For more direct control, you can switch me to Reactive Mode. When this mode is on, I will only act on comments where you specifically mention me with @jules. You can find this option in the Pull Request section of your global Jules UI settings. You can always switch back!

New to Jules? Learn more at jules.google/docs.


For security, I will only act on instructions from the user who triggered this task.

Copilot AI review requested due to automatic review settings July 8, 2026 18:21

Copilot AI left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Pull request overview

This PR updates user-facing copy and inline documentation to remove inaccurate claims that the Fireworks backend is “AMD-hosted,” and to emphasize that AMD proof-of-compute relies on self-hosted AMD hardware (e.g., MI300X/W7900) via vLLM + ROCm.

Changes:

  • Removed/softened “AMD-hosted Fireworks” wording across README, docs, deck text, and inline comments.
  • Clarified that AMD evidence should come from self-hosted MI300X/W7900 runs (not Fireworks).

Reviewed changes

Copilot reviewed 8 out of 8 changed files in this pull request and generated 3 comments.

Show a summary per file
File Description
README.md Removes “AMD-hosted” phrasing for Fireworks and updates backend descriptions.
gemmajudge/orchestrator.py Adjusts cost-report docstring to avoid “AMD-hosted Gemma” wording.
gemmajudge/config.py Updates backend enum comment to remove “AMD-hosted” reference for Fireworks.
Dockerfile Updates container comment to remove “Fireworks (AMD-hosted)” phrasing.
deck/build_deck.py Removes “Fireworks AMD-hosted infra” from slide strings and reframes AMD proof.
app.py Updates UI caption to restrict “Gemma/AMD evidence” to MI300X/W7900 runs.
AGENTS.md Removes “Fireworks AMD-hosted” claim from constraints/backends guidance.
.env.example Updates Fireworks section comment to remove “AMD-hosted” wording.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Comment thread README.md
Comment on lines 7 to +8
and a **Judge** that scores a target model's responses against a rubric — running the
whole closed loop on AMD (self-hosted on AMD GPUs via vLLM + ROCm, or Gemma on Fireworks'
AMD-hosted infrastructure).
whole closed loop on AMD (self-hosted on AMD GPUs via vLLM + ROCm, or Gemma on Fireworks).
Comment thread deck/build_deck.py
c.setFillColor(AMD)
c.setFont("Helvetica-Bold", 9)
c.drawString(x0, y - 0.28 * inch, "▲ both Gemma roles run on AMD (self-hosted via vLLM + ROCm, or Fireworks' AMD-hosted infra)")
c.drawString(x0, y - 0.28 * inch, "▲ both Gemma roles run on AMD (self-hosted via vLLM + ROCm)")
Comment thread AGENTS.md
Comment on lines 103 to 105
2. **AMD compute usage is REQUIRED or the project is disqualified.** Gemma must run on AMD:
self-hosted on MI300X (vLLM+ROCm) and/or Gemma on Fireworks' AMD-hosted infra. Proof
self-hosted on MI300X (vLLM+ROCm). Proof
(config + logs + screenshot) must be committed under `docs/amd_proof/` by end of Day 3.
Identified a contradiction where Fireworks was labeled as "AMD-hosted"
in multiple places (`README.md`, `AGENTS.md`, `app.py`, etc.), while
`DEPLOY.md` and `docs/real_runs/README.md` correctly state it runs on
NVIDIA H100 GPUs.

To comply with the "Honest copy: no fake AMD claims" design rule, removed
"AMD-hosted" descriptions for the Fireworks backend across the codebase.
Clarified that only the self-hosted MI300X/W7900 setups count as true
AMD compute proof.

Co-authored-by: Nevern1y <77607229+Nevern1y@users.noreply.github.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants