cardosofelipe/strix-halo-optimizations
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
25 Commits 1 Branch 0 Tags
474d94a07e2023027686925c0f4959bc5fa70d20
Commit Graph

7 Commits

Author SHA1 Message Date
Felipe Cardoso
474d94a07e chore: update model catalog with gemma 4, opus distill, and hw-bandwidth target 2026-04-03 20:03:53 +02:00
Felipe Cardoso
dd403a907c feat(serve): add optimized llama-server launcher with n-gram speculation 
Add `make serve` and `make serve-ngram` for launching llama-server with
baked-in optimal settings (Vulkan RADV, q4_0 KV cache, flash attention,
no-mmap, full GPU offload). N-gram speculative decoding gives 1.1-1.4x
tg speedup on repetitive content without upstream PR dependencies.
Update Phase 5 status: MTP is months away (4 unmerged PRs, no MoE
support), draft-model speculation stalled on ROCm buffer crashes.
 2026-03-30 21:12:30 +02:00
Felipe Cardoso
1549bc27c0 feat(optimize): add Phase 2 power profile and system tuning 
Add `make optimize-power` (ryzenadj 85W, sysctl, THP, RADV nogttspill)
with systemd services for boot/resume persistence. Integrate into
`make optimize --all` as Phase 2. Update optimization log with RyzenAdj
results (+46% tg at 70W sustained), KV sweep data, and quant shootout.
Add Qwen3-Coder-30B and Nemotron-Cascade-2 to model catalog.
 2026-03-30 18:53:52 +02:00
Felipe Cardoso
6f197a1455 fix: pass ARGS through in benchmark Makefile targets 
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
 2026-03-26 19:10:59 +01:00
Felipe Cardoso
58124cd657 feat: add Qwen3.5 model catalog and agentic evaluation framework 
Models:
- configs/models.conf: catalog with Qwen3.5-35B-A3B (MoE, top pick),
  Qwen3.5-27B (dense), Qwen3-Coder-30B-A3B (agentic/coding)
- Updated benchmark setup to show catalog with download status
- docs/model-recommendations.md: memory planning, quantization guide

Agentic evaluation:
- scripts/agentic/setup.sh: installs inspect-ai, evalplus, bigcodebench
  in a Python venv
- scripts/agentic/run-eval.sh: runs evaluations against local LLM server
  (ollama or llama.cpp). Suites: quick (HumanEval+IFEval), code
  (EvalPlus+BigCodeBench), tooluse (BFCL), full (all)
- bin/agentic: dispatcher with help
- docs/agentic-benchmarks.md: methodology, framework comparison, model
  recommendations for agentic use

Updated: Makefile (6 new targets), README, CLAUDE.md, docs/references.md

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
 2026-03-26 00:20:23 +01:00
Felipe Cardoso
a403dd9ce0 test: add BATS test suite (79 tests) 
- tests/common.bats: PROJECT_ROOT detection, is_cmd, timestamp, data_dir,
  logging functions, color handling, require_root
- tests/detect.bats: GPU sysfs reads with mock sysfs tree, kernel param
  parsing (word boundary, dot escaping, edge positions), recommended
  GTT/pages computation (64GB, 128GB, tiny, zero), firmware bad detection,
  stack detection
- tests/format.bats: human_bytes (0, KiB, MiB, GiB boundaries, 64GiB),
  human_mib (sub-GiB, exact-GiB, recommended values, empty input)
- tests/benchmark_compare.bats: improvement/regression display, empty
  results, missing files, usage output, config change detection
- tests/log_metrics.bats: CSV header, data format, field count, input
  validation, unknown argument handling
- tests/test_helper.sh: mock sysfs tree builder, bats-assert/support setup

Makefile: add 'make test' target

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
 2026-03-25 22:15:34 +01:00
Felipe Cardoso
c596e38e9e Initial commit 2026-03-25 20:13:15 +01:00