From ee0d1e12f257efc32edb24db037359d5b77ccc90 Mon Sep 17 00:00:00 2001 From: Lumpiasty Date: Sun, 31 May 2026 04:49:11 +0200 Subject: [PATCH] add llama-swap optimizations recommended by claude --- apps/llama/configs/config.yaml | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/apps/llama/configs/config.yaml b/apps/llama/configs/config.yaml index 7707125..9e8083a 100644 --- a/apps/llama/configs/config.yaml +++ b/apps/llama/configs/config.yaml @@ -4,16 +4,16 @@ logToStdout: "both" # proxy and upstream macros: base_args: "--no-warmup --port ${PORT} --mlock --no-mmap" - common_args: "--fit-target 1536 --no-warmup --port ${PORT} --no-mmap" + common_args: "--fit-target 256 --no-warmup --port ${PORT} --no-mmap -tb 12 -t 6" cpu_args: "--no-warmup --port ${PORT} -ngl 0" ctx_64k: "--ctx-size 65536" ctx_128k: "--ctx-size 131072" - ctx_256k: "--ctx-size 262144" - qwen35_think_args: "--temp 1.0 --top-p 0.95 --top-k 20 --min-p 0.00 -ctk q4_0 -ctv q4_0 --presence_penalty 1.5 --reasoning on" - qwen35_nothink_args: "--temp 0.7 --top-p 0.80 --top-k 20 --min-p 0.00 -ctk q4_0 -ctv q4_0 --presence_penalty 1.5 --reasoning off" + ctx_256k: "--ctx-size 131072" + qwen35_think_args: "--temp 1.0 --top-p 0.95 --top-k 20 --min-p 0.00 -ctk q8_0 -ctv q8_0 --presence_penalty 1.5 --reasoning on" + qwen35_nothink_args: "--temp 0.7 --top-p 0.80 --top-k 20 --min-p 0.00 -ctk q8_0 -ctv q8_0 --presence_penalty 1.5 --reasoning off" qwen35_35b_heretic_mmproj: "--mmproj-url https://huggingface.co/unsloth/Qwen3.5-35B-A3B-GGUF/resolve/main/mmproj-F16.gguf --mmproj /root/.cache/llama.cpp/unsloth_Qwen3.5-35B-A3B-GGUF_mmproj-F16.gguf" qwen35_4b_heretic_mmproj: "--mmproj-url https://huggingface.co/unsloth/Qwen3.5-4B-GGUF/resolve/main/mmproj-F16.gguf --mmproj /root/.cache/llama.cpp/unsloth_Qwen3.5-4B-GGUF_mmproj-F16.gguf" - gemma4_sampling: "--temp 1.0 --top-p 0.95 --top-k 64 -ctk q4_0 -ctv q4_0" + gemma4_sampling: "--temp 1.0 --top-p 0.95 --top-k 64 -ctk q8_0 -ctv q8_0" hooks: on_startup: