klaster/apps/llama/configs/config.yaml

# yaml-language-server: $schema=https://raw.githubusercontent.com/mostlygeek/llama-swap/refs/heads/main/config-schema.json
healthCheckTimeout: 600
logToStdout: "both" # proxy and upstream

macros:
  base_args: "--no-warmup --port ${PORT}"
  common_args: "--fit-target 1536 --no-warmup --port ${PORT}"
  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"
  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"

hooks:
  on_startup:
    preload:
      - "Qwen3.5-0.8B-GGUF-nothink:Q4_K_XL"

groups:
  always:
    persistent: true
    exclusive: false
    swap: false
    members:
      - "Qwen3.5-0.8B-GGUF-nothink:Q4_K_XL"

models:
  "Qwen3-Coder-Next-GGUF:Q4_K_M":
    cmd: |
      /app/llama-server
        -hf unsloth/Qwen3-Coder-Next-GGUF:Q4_K_M
        --ctx-size 65536
        --predict 8192
        --temp 1.0
        --min-p 0.01
        --top-p 0.95
        --top-k 40
        --repeat-penalty 1.0
        -ctk q4_0 -ctv q4_0
        ${common_args}

  "Qwen3.5-35B-A3B-GGUF:Q4_K_M":
    cmd: |
      /app/llama-server
        -hf unsloth/Qwen3.5-35B-A3B-GGUF:Q4_K_M
        ${ctx_256k}
        ${qwen35_think_args}
        ${common_args}

  "Qwen3.5-35B-A3B-GGUF-nothink:Q4_K_M":
    cmd: |
      /app/llama-server
        -hf unsloth/Qwen3.5-35B-A3B-GGUF:Q4_K_M
        ${ctx_256k}
        ${qwen35_nothink_args}
        ${common_args}

  # The "heretic" version does not provide the mmproj
  # so providing url to the one from the non-heretic version.
  "Qwen3.5-35B-A3B-heretic-GGUF:Q4_K_M":
    cmd: |
      /app/llama-server
        -hf mradermacher/Qwen3.5-35B-A3B-heretic-GGUF:Q4_K_M
        ${qwen35_35b_heretic_mmproj}
        ${ctx_256k}
        ${qwen35_think_args}
        ${common_args}

  "Qwen3.5-35B-A3B-heretic-GGUF-nothink:Q4_K_M":
    cmd: |
      /app/llama-server
        -hf mradermacher/Qwen3.5-35B-A3B-heretic-GGUF:Q4_K_M
        ${qwen35_35b_heretic_mmproj}
        ${ctx_256k}
        ${qwen35_nothink_args}
        ${common_args}

  "Qwen3.5-0.8B-GGUF-nothink:Q4_K_XL":
    cmd: |
      /app/llama-server
        -hf unsloth/Qwen3.5-0.8B-GGUF:Q4_K_XL
        --ctx-size 4096
        ${qwen35_nothink_args}
        ${base_args}

  "Qwen3.5-4B-GGUF:Q4_K_M":
    cmd: |
      /app/llama-server
        -hf unsloth/Qwen3.5-4B-GGUF:Q4_K_M
        ${ctx_128k}
        ${qwen35_think_args}
        ${common_args}

  "Qwen3.5-4B-GGUF-nothink:Q4_K_M":
    cmd: |
      /app/llama-server
        -hf unsloth/Qwen3.5-4B-GGUF:Q4_K_M
        ${ctx_128k}
        ${qwen35_nothink_args}
        ${common_args}

  "Qwen3.5-4B-heretic-GGUF:Q4_K_M":
    cmd: |
      /app/llama-server
        -hf mradermacher/Qwen3.5-4B-heretic-GGUF:Q4_K_M
        ${qwen35_4b_heretic_mmproj}
        ${ctx_128k}
        ${qwen35_think_args}
        ${common_args}

  "Qwen3.5-4B-heretic-GGUF-nothink:Q4_K_M":
    cmd: |
      /app/llama-server
        -hf mradermacher/Qwen3.5-4B-heretic-GGUF:Q4_K_M
        ${qwen35_4b_heretic_mmproj}
        ${ctx_128k}
        ${qwen35_nothink_args}
        ${common_args}

  "gemma-4-26B-A4B-it:UD-Q4_K_XL":
    cmd: |
      /app/llama-server
        -hf unsloth/gemma-4-26B-A4B-it-GGUF:UD-Q4_K_XL \
        ${ctx_256k}
        ${gemma4_sampling}
        ${common_args}

  "gemma-4-26B-A4B-it:UD-Q2_K_XL":
    cmd: |
      /app/llama-server
        -hf unsloth/gemma-4-26B-A4B-it-GGUF:UD-Q2_K_XL \
        ${ctx_256k}
        ${gemma4_sampling}
        ${common_args}

  "unsloth/gemma-4-E4B-it-GGUF:UD-Q4_K_XL":
    cmd: |
      /app/llama-server
        -hf unsloth/gemma-4-E4B-it-GGUF:UD-Q4_K_XL \
        ${ctx_128k}
        ${gemma4_sampling}
        ${common_args}

  "unsloth/gemma-4-E2B-it-GGUF:UD-Q4_K_XL":
    cmd: |
      /app/llama-server
        -hf unsloth/gemma-4-E2B-it-GGUF:UD-Q4_K_XL \
        ${ctx_128k}
        ${gemma4_sampling}
        ${common_args}

  "unsloth/Qwen3.6-35B-A3B-GGUF:UD-Q4_K_XL":
    cmd: |
      /app/llama-server
      -hf unsloth/Qwen3.6-35B-A3B-GGUF:UD-Q4_K_XL
      ${ctx_256k}
      ${qwen35_think_args}
      ${common_args}

  "unsloth/Qwen3.6-35B-A3B-GGUF-nothink:UD-Q4_K_XL":
    cmd: |
      /app/llama-server
      -hf unsloth/Qwen3.6-35B-A3B-GGUF:UD-Q4_K_XL
      ${ctx_256k}
      ${qwen35_nothink_args}
      ${common_args}