DEDUPED Hot AI repos, turned into use cases Get the Monday 5

How to get repeatable model outputs for regression tests with Deep Infer

Use batch-invariant inference and a fixed seed so identical inputs give identical outputs.

Who it is forA QA or ML engineer who needs to check whether a model change alters outputs and must rule out batch-related noise.

The problemOutputs can change with batch composition and sampling randomness, which makes before and after comparisons unreliable.

The outcomeA test setup where the same input tokens and sampling parameters return the same output on every run on NVIDIA hardware.

What you need first

  • Linux with GCC 13.0 or newer and Python 3.12
  • NVIDIA SM100 GPU with CUDA 13, or Huawei Ascend 950 NPU with CANN 9.2
  • Access to the deepseek-ai/DeepSeek-V4.1-Flash-1017 checkpoint on Hugging Face, or a local copy
  • Enough disk space and memory for the checkpoint and the build

Steps

  1. Run DEEPINFER_BATCH_INVARIANT=1 DEEPINFER_ENABLE_MTP_VARLEN=0 in the shell that runs your inference script.
  2. Pass a fixed seq_seed to SamplingParams in your script.
  3. Run the same prompts twice with the same settings and save both outputs.
  4. Compare the two saved outputs and record any difference as a failed check.

Prompt for your agent

In the DeepInfer repo, set DEEPINFER_BATCH_INVARIANT=1 and DEEPINFER_ENABLE_MTP_VARLEN=0, pass a fixed seq_seed to SamplingParams, and run <prompt file> twice. Output a diff of the two result files. Verify that the diff is empty.

Caveat. The README says both batch-invariant inference and deterministic sampling are NVIDIA only, so this setup does not apply to Ascend.

Use this in your AI agent

Ask your agent

Using the Deduped MCP, get the setup steps for deepseek-ai/DeepInfer and "Get repeatable model outputs for regression tests"

Add Deduped to Claude Code once

claude mcp add --transport http deduped https://dedupedai.com/mcp

Cursor, ChatGPT, others

Share: X LinkedIn Hacker News Email Copy link

Third-party repos we don't own. AI-written use cases can be wrong. Disclaimer