Skip to content
C-Kernel-Engine docs Antshiv Open Source

C-KERNEL-ENGINE · GUIDE

Gemma4 speculative assistant

Capability matrix

  • Compute and format Matching draft and backbone · See paired run record.
  • Local entry point Pair validation.
  • Server Unverified for this artifact. Paired inference
  • Harness No recorded coding task. Validate the pair together before measuring speed or serving it.

When to use it

A paired draft model proposes tokens for a matching Gemma4 backbone. Use the published pair guide to check weights and validation before trying it. A standalone chat command is inappropriate here.

Configuration and tested scope

Hardware and tuning

No comparable x86/ARM prefill and decode benchmark pair is recorded here. Full-checkpoint training, LoRA and other PEFT methods: no verified CKE recipe for this exact checkpoint in the reviewed evidence. For a trainable starter, use the FP32 and adapter recipes.

Prepare the two bundles

Use the Google assistant checkpoint alongside the matching E4B backbone. These are preparation commands from the Xeon validation plan.

Examplebash
.venv/bin/python version/v8/scripts/ck_run_v8.py run \
  hf://unsloth/gemma-4-E4B-it-GGUF/gemma-4-E4B-it-Q4_K_M.gguf \
  --run /tmp/ck-gemma4-e4b-runtime \
  --context-len 2048 \
  --force-convert --force-compile \
  --prompt 'Give me a detailed example of C code.' \
  --chat-template gemma4 \
  --max-tokens 64 \
  --temperature 0.0

.venv/bin/python version/v8/scripts/ck_run_v8.py run \
  hf://google/gemma-4-E4B-it-assistant \
  --run /tmp/ck-gemma4-assistant-runtime \
  --context-len 2048 \
  --force-convert --force-compile \
  --generate-only \
  --chat-template none \
  --allow-raw-prompt