swap: true doesn't do anything

[Garaymie]

If I make 3 requests at the same time to different model in group1 (for example: gemma-3-27b-it, Mistral-Small-24B-Instruct-2501, EtherealAurora-12B-v2) it will load them simultaneously, but I have them in group1 with swap: true that should not allow to load them simultaneously. What am I doing wrong?

My config:

healthCheckTimeout: 1800

models:
  "gemma-3-27b-it":
    cmd: |
      llama-server
      -m /models/gemma-3-27b-it-UD-IQ2_M.gguf
      --ctx-size 2048
      --cache-type-k q8_0
      --cache-type-v q8_0
      --flash-attn
      --mlock
      --port ${PORT}
  "Mistral-Small-24B-Instruct-2501":
    cmd: |
      ik_llama-server
      -m /models/Mistral-Small-24B-Instruct-2501.i1-IQ3_XXS.gguf
      --ctx-size 2048
      --cache-type-k q8_0
      --cache-type-v q8_0
      --mlock
      --gpu-layers 0
      --temp 0.15
      --top-p 1.00
      --flash-attn
      --port ${PORT}
  "EtherealAurora-12B-v2":
    cmd: |
      llama-server
      -m /models/EtherealAurora-12B-v2.i1-Q4_K_M.gguf
      --ctx-size 2048
      --cache-type-k q8_0
      --cache-type-v q8_0
      --mlock
      --gpu-layers 0
      --flash-attn
      --port ${PORT}
  "solar-10.7b-instruct-v1.0":
    cmd: |
      llama-server
      -m /models/solar-10.7b-instruct-v1.0.Q4_K_M.gguf
      --ctx-size 2048
      --cache-type-k q8_0
      --cache-type-v q8_0
      --flash-attn
      --port ${PORT}
  "TiTan-Qwen2.5-0.5B":
    cmd: |
      llama-server
      -m /models/TiTan-Qwen2.5-0.5B-q4_k_m.gguf
      --ctx-size 2048
      --cache-type-k q8_0
      --cache-type-v q8_0
      --temp 0.7
      --top-p 0.9
      --flash-attn
      --keep -1
      --port ${PORT}

groups:
  group1:
    swap: true
    exclusive: false
    members:
      - gemma-3-27b-it
      - Mistral-Small-24B-Instruct-2501
      - EtherealAurora-12B-v2
      - solar-10.7b-instruct-v1.0
  group2:
    swap: false
    exclusive: false
    members:
      - TiTan-Gemma3-0.27B

[mostlygeek]

Hi, set logging level to debug and share the proxy logs when you’re requesting the models simultaneously please.

[Garaymie]

Here:

llama-swap listening on :2345
[INFO] Request 192.168.128.3 "GET /api/version HTTP/1.1" 200 19 "Python/3.11 aiohttp/3.11.11" 52.007µs
[INFO] Request 192.168.128.3 "GET /api/version HTTP/1.1" 200 19 "Python/3.11 aiohttp/3.11.11" 79.78µs
[INFO] Request 192.168.128.3 "GET /api/version HTTP/1.1" 200 19 "Python/3.11 aiohttp/3.11.11" 25.327µs
[DEBUG] <Mistral-Small-24B-Instruct-2501> swapState() State transitioned from stopped to starting
[DEBUG] <Mistral-Small-24B-Instruct-2501> Executing start command: ik_llama-server -m /models/Mistral-Small-24B-Instruct-2501.i1-IQ3_XXS.gguf --ctx-size 2048 --cache-type-k q8_0 --cache-type-v q8_0 --mlock --gpu-layers 0 --temp 0.15 --top-p 1.00 --flash-attn --port 5802, env: 
[DEBUG] <EtherealAurora-12B-v2> swapState() State transitioned from stopped to starting
[DEBUG] <EtherealAurora-12B-v2> Executing start command: llama-server -m /models/EtherealAurora-12B-v2.i1-Q4_K_M.gguf --ctx-size 2048 --cache-type-k q8_0 --cache-type-v q8_0 --mlock --gpu-layers 0 --flash-attn --port 5800, env: 
[DEBUG] <gemma-3-27b-it> swapState() State transitioned from stopped to starting
[DEBUG] <gemma-3-27b-it> Executing start command: llama-server -m /models/gemma-3-27b-it-UD-IQ2_M.gguf --ctx-size 2048 --cache-type-k q8_0 --cache-type-v q8_0 --flash-attn --mlock --port 5808, env: 
INFO [                    main] build info | tid="140543158959488" timestamp=1756327762 build=0 commit="unknown"
INFO [                    main] system info | tid="140543158959488" timestamp=1756327762 n_threads=6 n_threads_batch=-1 total_threads=12 system_info="AVX = 1 | AVX_VNNI = 0 | AVX2 = 1 | AVX512 = 0 | AVX512_VBMI = 0 | AVX512_VNNI = 0 | AVX512_BF16 = 0 | FMA = 1 | NEON = 0 | SVE = 0 | ARM_FMA = 0 | F16C = 1 | FP16_VA = 0 | WASM_SIMD = 0 | BLAS = 1 | SSE3 = 1 | SSSE3 = 1 | VSX = 0 | MATMUL_INT8 = 0 | LLAMAFILE = 1 | "
llama_model_loader: loaded meta data with 52 key-value pairs and 363 tensors from /models/Mistral-Small-24B-Instruct-2501.i1-IQ3_XXS.gguf (version GGUF V3 (latest))
llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output.
llama_model_loader: - kv   0:                       general.architecture str              = llama
llama_model_loader: - kv   1:                               general.type str              = model
llama_model_loader: - kv   2:                               general.name str              = Mistral Small 24B Instruct 2501
llama_model_loader: - kv   3:                            general.version str              = 2501
llama_model_loader: - kv   4:                           general.finetune str              = Instruct
llama_model_loader: - kv   5:                           general.basename str              = Mistral-Small
llama_model_loader: - kv   6:                         general.size_label str              = 24B
llama_model_loader: - kv   7:                            general.license str              = apache-2.0
llama_model_loader: - kv   8:                   general.base_model.count u32              = 1
llama_model_loader: - kv   9:                  general.base_model.0.name str              = Mistral Small 24B Instruct 2501
llama_model_loader: - kv  10:               general.base_model.0.version str              = 2501
llama_model_loader: - kv  11:          general.base_model.0.organization str              = Mistralai
llama_model_loader: - kv  12:              general.base_model.0.repo_url str              = https://huggingface.co/mistralai/Mist...
llama_model_loader: - kv  13:                               general.tags arr[str,3]       = ["transformer"]
llama_model_loader: - kv  14:                          general.languages arr[str,10]      = ["en", "fr", "de", "es", "it", "pt", ...
llama_model_loader: - kv  15:                          llama.block_count u32              = 40
llama_model_loader: - kv  16:                       llama.context_length u32              = 32768
llama_model_loader: - kv  17:                     llama.embedding_length u32              = 5120
llama_model_loader: - kv  18:                  llama.feed_forward_length u32              = 32768
llama_model_loader: - kv  19:                 llama.attention.head_count u32              = 32
llama_model_loader: - kv  20:              llama.attention.head_count_kv u32              = 8
llama_model_loader: - kv  21:                       llama.rope.freq_base f32              = 100000000.000000
llama_model_loader: - kv  22:     llama.attention.layer_norm_rms_epsilon f32              = 0.000010
llama_model_loader: - kv  23:                 llama.attention.key_length u32              = 128
llama_model_loader: - kv  24:               llama.attention.value_length u32              = 128
llama_model_loader: - kv  25:                           llama.vocab_size u32              = 131072
llama_model_loader: - kv  26:                 llama.rope.dimension_count u32              = 128
llama_model_loader: - kv  27:                       tokenizer.ggml.model str              = gpt2
llama_model_loader: - kv  28:                         tokenizer.ggml.pre str              = tekken
llama_model_loader: - kv  29:                      tokenizer.ggml.tokens arr[str,131072]  = ["<unk>", "<s>", "</s>", "[INST]", "[...
llama_model_loader: - kv  30:                  tokenizer.ggml.token_type arr[i32,131072]  = [3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, ...
llama_model_loader: - kv  31:                      tokenizer.ggml.merges arr[str,269443]  = ["Ġ Ġ", "Ġ t", "e r", "i n", "Ġ �...
llama_model_loader: - kv  32:                tokenizer.ggml.bos_token_id u32              = 1
llama_model_loader: - kv  33:                tokenizer.ggml.eos_token_id u32              = 2
llama_model_loader: - kv  34:            tokenizer.ggml.unknown_token_id u32              = 0
llama_model_loader: - kv  35:               tokenizer.ggml.add_bos_token bool             = true
llama_model_loader: - kv  36:               tokenizer.ggml.add_eos_token bool             = false
llama_model_loader: - kv  37:                    tokenizer.chat_template str              = {%- set today = strftime_now("%Y-%m-%...
llama_model_loader: - kv  38:            tokenizer.ggml.add_space_prefix bool             = false
llama_model_loader: - kv  39:               general.quantization_version u32              = 2
llama_model_loader: - kv  40:                          general.file_type u32              = 23
llama_model_loader: - kv  41:                                general.url str              = https://huggingface.co/mradermacher/M...
llama_model_loader: - kv  42:              mradermacher.quantize_version str              = 2
llama_model_loader: - kv  43:                  mradermacher.quantized_by str              = mradermacher
llama_model_loader: - kv  44:                  mradermacher.quantized_at str              = 2025-02-04T03:27:30+01:00
llama_model_loader: - kv  45:                  mradermacher.quantized_on str              = marco
llama_model_loader: - kv  46:                         general.source.url str              = https://huggingface.co/huihui-ai/Mist...
llama_model_loader: - kv  47:                  mradermacher.convert_type str              = hf
llama_model_loader: - kv  48:                      quantize.imatrix.file str              = Mistral-Small-24B-Instruct-2501.i1-IQ...
llama_model_loader: - kv  49:                   quantize.imatrix.dataset str              = imatrix-training-full-3
llama_model_loader: - kv  50:             quantize.imatrix.entries_count i32              = 280
llama_model_loader: - kv  51:              quantize.imatrix.chunks_count i32              = 321
llama_model_loader: - type  f32:   81 tensors
llama_model_loader: - type q4_K:   40 tensors
llama_model_loader: - type q5_K:    1 tensors
llama_model_loader: - type iq3_xxs:  120 tensors
llama_model_loader: - type iq3_s:   41 tensors
llama_model_loader: - type iq2_s:   80 tensors
[DEBUG] <gemma-3-27b-it> Connection refused on http://localhost:5808/health, giving up in 1800s (normal during startup)
[DEBUG] <EtherealAurora-12B-v2> Connection refused on http://localhost:5800/health, giving up in 1800s (normal during startup)
ggml_vulkan: Found 1 Vulkan devices:
ggml_vulkan: Found 1 Vulkan devices:
ggml_vulkan: 0 = AMD Radeon Graphics (RADV RENOIR) (radv) | uma: 1 | fp16: 1 | bf16: 0 | warp size: 64 | shared memory: 65536 | int dot: 0 | matrix cores: none
ggml_vulkan: 0 = AMD Radeon Graphics (RADV RENOIR) (radv) | uma: 1 | fp16: 1 | bf16: 0 | warp size: 64 | shared memory: 65536 | int dot: 0 | matrix cores: none
build: 1 (b730706) with cc (GCC) 15.2.1 20250813 for x86_64-pc-linux-gnu
system info: n_threads = 6, n_threads_batch = 6, total_threads = 12

system_info: n_threads = 6 (n_threads_batch = 6) / 12 | CPU : SSE3 = 1 | SSSE3 = 1 | AVX = 1 | AVX2 = 1 | F16C = 1 | FMA = 1 | BMI2 = 1 | LLAMAFILE = 1 | OPENMP = 1 | REPACK = 1 | 

main: binding port with default address family
build: 1 (b730706) with cc (GCC) 15.2.1 20250813 for x86_64-pc-linux-gnu
system info: n_threads = 6, n_threads_batch = 6, total_threads = 12

system_info: n_threads = 6 (n_threads_batch = 6) / 12 | CPU : SSE3 = 1 | SSSE3 = 1 | AVX = 1 | AVX2 = 1 | F16C = 1 | FMA = 1 | BMI2 = 1 | LLAMAFILE = 1 | OPENMP = 1 | REPACK = 1 | 

main: binding port with default address family
main: HTTP server is listening, hostname: 127.0.0.1, port: 5800, http threads: 11
main: loading model
srv    load_model: loading model '/models/EtherealAurora-12B-v2.i1-Q4_K_M.gguf'
llama_model_load_from_file_impl: using device Vulkan0 (AMD Radeon Graphics (RADV RENOIR)) - 5802 MiB free
main: HTTP server is listening, hostname: 127.0.0.1, port: 5808, http threads: 11
main: loading model
srv    load_model: loading model '/models/gemma-3-27b-it-UD-IQ2_M.gguf'
llama_model_load_from_file_impl: using device Vulkan0 (AMD Radeon Graphics (RADV RENOIR)) - 5802 MiB free
llama_model_loader: loaded meta data with 40 key-value pairs and 808 tensors from /models/gemma-3-27b-it-UD-IQ2_M.gguf (version GGUF V3 (latest))
llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output.
llama_model_loader: - kv   0:                       general.architecture str              = gemma3
llama_model_loader: - kv   1:                               general.type str              = model
llama_model_loader: - kv   2:                               general.name str              = Gemma-3-27B-It
llama_model_loader: - kv   3:                           general.finetune str              = it
llama_model_loader: - kv   4:                           general.basename str              = Gemma-3-27B-It
llama_model_loader: - kv   5:                       general.quantized_by str              = Unsloth
llama_model_loader: - kv   6:                         general.size_label str              = 27B
llama_model_loader: - kv   7:                           general.repo_url str              = https://huggingface.co/unsloth
llama_model_loader: - kv   8:                      gemma3.context_length u32              = 131072
llama_model_loader: - kv   9:                    gemma3.embedding_length u32              = 5376
llama_model_loader: - kv  10:                         gemma3.block_count u32              = 62
llama_model_loader: - kv  11:                 gemma3.feed_forward_length u32              = 21504
llama_model_loader: - kv  12:                gemma3.attention.head_count u32              = 32
llama_model_loader: - kv  13:    gemma3.attention.layer_norm_rms_epsilon f32              = 0.000001
llama_model_loader: - kv  14:                gemma3.attention.key_length u32              = 128
llama_model_loader: - kv  15:              gemma3.attention.value_length u32              = 128
llama_model_loader: - kv  16:                      gemma3.rope.freq_base f32              = 1000000.000000
llama_model_loader: - kv  17:            gemma3.attention.sliding_window u32              = 1024
llama_model_loader: - kv  18:             gemma3.attention.head_count_kv u32              = 16
llama_model_loader: - kv  19:                   gemma3.rope.scaling.type str              = linear
llama_model_loader: - kv  20:                 gemma3.rope.scaling.factor f32              = 8.000000
llama_model_loader: - kv  21:                       tokenizer.ggml.model str              = llama
llama_model_loader: - kv  22:                         tokenizer.ggml.pre str              = default
llama_model_loader: loaded meta data with 55 key-value pairs and 363 tensors from /models/EtherealAurora-12B-v2.i1-Q4_K_M.gguf (version GGUF V3 (latest))
llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output.
llama_model_loader: - kv   0:                       general.architecture str              = llama
llama_model_loader: - kv   1:                               general.type str              = model
llama_model_loader: - kv   2:                               general.name str              = EtherealAurora 12B
llama_model_loader: - kv   3:                            general.version str              = v2
llama_model_loader: - kv   4:                       general.organization str              = Yamatazen
llama_model_loader: - kv   5:                           general.basename str              = EtherealAurora
llama_model_loader: - kv   6:                         general.size_label str              = 12B
llama_model_loader: - kv   7:                   general.base_model.count u32              = 2
llama_model_loader: - kv   8:                  general.base_model.0.name str              = EtherealAurora 12B
llama_model_loader: - kv   9:          general.base_model.0.organization str              = Yamatazen
llama_model_loader: - kv  10:              general.base_model.0.repo_url str              = https://huggingface.co/yamatazen/Ethe...
llama_model_loader: - kv  11:                  general.base_model.1.name str              = Mistral Nemo Instruct 2407
llama_model_loader: - kv  12:               general.base_model.1.version str              = 2407
llama_model_loader: - kv  13:          general.base_model.1.organization str              = Natong19
llama_model_loader: - kv  14:              general.base_model.1.repo_url str              = https://huggingface.co/natong19/Mistr...
llama_model_loader: - kv  15:                               general.tags arr[str,3]       = ["mergekit", "merge", "chatml"]
llama_model_loader: - kv  16:                          general.languages arr[str,2]       = ["en", "ja"]
llama_model_loader: - kv  17:                          llama.block_count u32              = 40
llama_model_loader: - kv  18:                       llama.context_length u32              = 1024000
llama_model_loader: - kv  19:                     llama.embedding_length u32              = 5120
llama_model_loader: - kv  20:                  llama.feed_forward_length u32              = 14336
llama_model_loader: - kv  21:                 llama.attention.head_count u32              = 32
llama_model_loader: - kv  22:              llama.attention.head_count_kv u32              = 8
llama_model_loader: - kv  23:                       llama.rope.freq_base f32              = 1000000.000000
llama_model_loader: - kv  24:     llama.attention.layer_norm_rms_epsilon f32              = 0.000010
llama_model_loader: - kv  25:                 llama.attention.key_length u32              = 128
llama_model_loader: - kv  26:               llama.attention.value_length u32              = 128
llama_model_loader: - kv  27:                           llama.vocab_size u32              = 131072
llama_model_loader: - kv  28:                 llama.rope.dimension_count u32              = 128
llama_model_loader: - kv  29:                       tokenizer.ggml.model str              = gpt2
llama_model_loader: - kv  30:                         tokenizer.ggml.pre str              = tekken
llama_model_loader: - kv  31:                      tokenizer.ggml.tokens arr[str,131072]  = ["<unk>", "<s>", "<|im_end|>", "<|im_...
llama_model_loader: - kv  23:                      tokenizer.ggml.tokens arr[str,262208]  = ["<pad>", "<eos>", "<bos>", "<unk>", ...
llama_model_loader: - kv  32:                  tokenizer.ggml.token_type arr[i32,131072]  = [3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, ...
llama_model_loader: - kv  33:                      tokenizer.ggml.merges arr[str,269443]  = ["Ġ Ġ", "Ġ t", "e r", "i n", "Ġ �...
llama_model_loader: - kv  34:                tokenizer.ggml.bos_token_id u32              = 1
llama_model_loader: - kv  35:                tokenizer.ggml.eos_token_id u32              = 2
llama_model_loader: - kv  36:            tokenizer.ggml.unknown_token_id u32              = 0
llama_model_loader: - kv  37:            tokenizer.ggml.padding_token_id u32              = 10
llama_model_loader: - kv  38:               tokenizer.ggml.add_bos_token bool             = true
llama_model_loader: - kv  39:               tokenizer.ggml.add_eos_token bool             = false
llama_model_loader: - kv  40:                    tokenizer.chat_template str              = {% if not add_generation_prompt is de...
llama_model_loader: - kv  41:            tokenizer.ggml.add_space_prefix bool             = false
llama_model_loader: - kv  42:               general.quantization_version u32              = 2
llama_model_loader: - kv  43:                          general.file_type u32              = 15
llama_model_loader: - kv  44:                                general.url str              = https://huggingface.co/mradermacher/E...
llama_model_loader: - kv  45:              mradermacher.quantize_version str              = 2
llama_model_loader: - kv  46:                  mradermacher.quantized_by str              = mradermacher
llama_model_loader: - kv  47:                  mradermacher.quantized_at str              = 2025-03-12T12:46:50+01:00
llama_model_loader: - kv  48:                  mradermacher.quantized_on str              = marco
llama_model_loader: - kv  49:                         general.source.url str              = https://huggingface.co/yamatazen/Ethe...
llama_model_loader: - kv  50:                  mradermacher.convert_type str              = hf
llama_model_loader: - kv  51:                      quantize.imatrix.file str              = EtherealAurora-12B-v2-i1-GGUF/imatrix...
llama_model_loader: - kv  52:                   quantize.imatrix.dataset str              = imatrix-training-full-3
llama_model_loader: - kv  53:             quantize.imatrix.entries_count i32              = 280
llama_model_loader: - kv  54:              quantize.imatrix.chunks_count i32              = 321
llama_model_loader: - type  f32:   81 tensors
llama_model_loader: - type q4_K:  241 tensors
llama_model_loader: - type q6_K:   41 tensors
print_info: file format = GGUF V3 (latest)
print_info: file type   = Q4_K - Medium
print_info: file size   = 6.96 GiB (4.88 BPW) 
init_tokenizer: initializing tokenizer for type 2
llama_model_loader: - kv  24:                      tokenizer.ggml.scores arr[f32,262208]  = [-1000.000000, -1000.000000, -1000.00...
llama_model_loader: - kv  25:                  tokenizer.ggml.token_type arr[i32,262208]  = [3, 3, 3, 3, 3, 4, 3, 3, 3, 3, 3, 3, ...
llama_model_loader: - kv  26:                tokenizer.ggml.bos_token_id u32              = 2
llama_model_loader: - kv  27:                tokenizer.ggml.eos_token_id u32              = 106
llama_model_loader: - kv  28:            tokenizer.ggml.unknown_token_id u32              = 3
llama_model_loader: - kv  29:            tokenizer.ggml.padding_token_id u32              = 0
llama_model_loader: - kv  30:               tokenizer.ggml.add_bos_token bool             = true
llama_model_loader: - kv  31:               tokenizer.ggml.add_eos_token bool             = false
llama_model_loader: - kv  32:                    tokenizer.chat_template str              = {{ bos_token }}\n{%- if messages[0]['r...
llama_model_loader: - kv  33:            tokenizer.ggml.add_space_prefix bool             = false
llama_model_loader: - kv  34:               general.quantization_version u32              = 2
llama_model_loader: - kv  35:                          general.file_type u32              = 29
llama_model_loader: - kv  36:                      quantize.imatrix.file str              = gemma-3-27b-it-GGUF/imatrix_unsloth.dat
llama_model_loader: - kv  37:                   quantize.imatrix.dataset str              = unsloth_calibration_gemma-3-27b-it.txt
llama_model_loader: - kv  38:             quantize.imatrix.entries_count i32              = 434
llama_model_loader: - kv  39:              quantize.imatrix.chunks_count i32              = 663
llama_model_loader: - type  f32:  373 tensors
llama_model_loader: - type q5_K:    1 tensors
llama_model_loader: - type iq2_xs:    5 tensors
llama_model_loader: - type iq3_xxs:   20 tensors
llama_model_loader: - type iq3_s:  135 tensors
llama_model_loader: - type iq2_s:  274 tensors
print_info: file format = GGUF V3 (latest)
print_info: file type   = IQ2_M - 2.7 bpw
print_info: file size   = 8.96 GiB (2.85 BPW) 
load: special_eos_id is not in special_eog_ids - the tokenizer config may be incorrect
load: printing all EOG tokens:
load:   - 2 ('</s>')
load: special tokens cache size = 1000
load: token to piece cache size = 0.8498 MB
llm_load_print_meta: format           = GGUF V3 (latest)
llm_load_print_meta: arch             = llama
llm_load_print_meta: n_ctx_train      = 32768
llm_load_print_meta: n_embd           = 5120
llm_load_print_meta: n_layer          = 40
llm_load_print_meta: n_head           = 32
llm_load_print_meta: n_head_kv        = 8
llm_load_print_meta: n_rot            = 128
llm_load_print_meta: n_swa            = 0
llm_load_print_meta: n_swa_pattern    = 1
llm_load_print_meta: n_embd_head_k    = 128
llm_load_print_meta: n_embd_head_v    = 128
llm_load_print_meta: n_gqa            = 4
llm_load_print_meta: n_embd_k_gqa     = 1024
llm_load_print_meta: n_embd_v_gqa     = 1024
llm_load_print_meta: f_norm_eps       = 0.0e+00
llm_load_print_meta: f_norm_rms_eps   = 1.0e-05
llm_load_print_meta: f_clamp_kqv      = 0.0e+00
llm_load_print_meta: f_max_alibi_bias = 0.0e+00
llm_load_print_meta: f_logit_scale    = 0.0e+00
llm_load_print_meta: n_ff             = 32768
llm_load_print_meta: n_expert         = 0
llm_load_print_meta: n_expert_used    = 0
llm_load_print_meta: causal attn      = 1
llm_load_print_meta: pooling type     = 0
llm_load_print_meta: rope type        = 0
llm_load_print_meta: rope scaling     = linear
llm_load_print_meta: freq_base_train  = 100000000.0
llm_load_print_meta: freq_scale_train = 1
llm_load_print_meta: n_ctx_orig_yarn  = 32768
llm_load_print_meta: rope_finetuned   = unknown
llm_load_print_meta: ssm_d_conv       = 0
llm_load_print_meta: ssm_d_inner      = 0
llm_load_print_meta: ssm_d_state      = 0
llm_load_print_meta: ssm_dt_rank      = 0
llm_load_print_meta: model type       = 13B
llm_load_print_meta: model ftype      = IQ3_XXS - 3.0625 bpw
llm_load_print_meta: model params     = 23.572 B
llm_load_print_meta: model size       = 8.636 GiB (3.147 BPW) 
llm_load_print_meta: repeating layers = 7.938 GiB (3.067 BPW, 22.230 B parameters)
llm_load_print_meta: general.name     = Mistral Small 24B Instruct 2501
print_info: vocab type       = BPE
print_info: n_vocab          = 131072
print_info: n_merges         = 269443
print_info: BOS token        = 1 '<s>'
print_info: EOS token        = 2 '</s>'
print_info: UNK token        = 0 '<unk>'
print_info: LF token         = 1010 'Ċ'
print_info: EOG token        = 2 '</s>'
print_info: max token length = 150
llm_load_tensors: ggml ctx size =    0.17 MiB
load: printing all EOG tokens:
load:   - 2 ('<|im_end|>')
load: special tokens cache size = 1000
load: token to piece cache size = 0.8499 MB
print_info: arch             = llama
print_info: vocab_only       = 0
print_info: n_ctx_train      = 1024000
print_info: n_embd           = 5120
print_info: n_layer          = 40
print_info: n_head           = 32
print_info: n_head_kv        = 8
print_info: n_rot            = 128
print_info: n_swa            = 0
print_info: is_swa_any       = 0
print_info: n_embd_head_k    = 128
print_info: n_embd_head_v    = 128
print_info: n_gqa            = 4
print_info: n_embd_k_gqa     = 1024
print_info: n_embd_v_gqa     = 1024
print_info: f_norm_eps       = 0.0e+00
print_info: f_norm_rms_eps   = 1.0e-05
print_info: f_clamp_kqv      = 0.0e+00
print_info: f_max_alibi_bias = 0.0e+00
print_info: f_logit_scale    = 0.0e+00
print_info: f_attn_scale     = 0.0e+00
print_info: n_ff             = 14336
print_info: n_expert         = 0
print_info: n_expert_used    = 0
print_info: causal attn      = 1
print_info: pooling type     = 0
print_info: rope type        = 0
print_info: rope scaling     = linear
print_info: freq_base_train  = 1000000.0
print_info: freq_scale_train = 1
print_info: n_ctx_orig_yarn  = 1024000
print_info: rope_finetuned   = unknown
print_info: model type       = 13B
print_info: model params     = 12.25 B
print_info: general.name     = EtherealAurora 12B
print_info: vocab type       = BPE
print_info: n_vocab          = 131072
print_info: n_merges         = 269443
print_info: BOS token        = 1 '<s>'
print_info: EOS token        = 2 '<|im_end|>'
print_info: EOT token        = 2 '<|im_end|>'
print_info: UNK token        = 0 '<unk>'
print_info: PAD token        = 10 '<pad>'
print_info: LF token         = 1010 'Ċ'
print_info: EOG token        = 2 '<|im_end|>'
print_info: max token length = 150
load_tensors: loading model tensors, this can take a while... (mmap = true)
load: printing all EOG tokens:
load:   - 106 ('<end_of_turn>')
load: special tokens cache size = 6415
load: token to piece cache size = 1.9446 MB
print_info: arch             = gemma3
print_info: vocab_only       = 0
print_info: n_ctx_train      = 131072
print_info: n_embd           = 5376
print_info: n_layer          = 62
print_info: n_head           = 32
print_info: n_head_kv        = 16
print_info: n_rot            = 128
print_info: n_swa            = 1024
print_info: is_swa_any       = 1
print_info: n_embd_head_k    = 128
print_info: n_embd_head_v    = 128
print_info: n_gqa            = 2
print_info: n_embd_k_gqa     = 2048
print_info: n_embd_v_gqa     = 2048
print_info: f_norm_eps       = 0.0e+00
print_info: f_norm_rms_eps   = 1.0e-06
print_info: f_clamp_kqv      = 0.0e+00
print_info: f_max_alibi_bias = 0.0e+00
print_info: f_logit_scale    = 0.0e+00
print_info: f_attn_scale     = 7.7e-02
print_info: n_ff             = 21504
print_info: n_expert         = 0
print_info: n_expert_used    = 0
print_info: causal attn      = 1
print_info: pooling type     = 0
print_info: rope type        = 2
print_info: rope scaling     = linear
print_info: freq_base_train  = 1000000.0
print_info: freq_scale_train = 0.125
print_info: n_ctx_orig_yarn  = 131072
print_info: rope_finetuned   = unknown
print_info: model type       = 27B
print_info: model params     = 27.01 B
print_info: general.name     = Gemma-3-27B-It
print_info: vocab type       = SPM
print_info: n_vocab          = 262208
print_info: n_merges         = 0
print_info: BOS token        = 2 '<bos>'
print_info: EOS token        = 106 '<end_of_turn>'
print_info: EOT token        = 106 '<end_of_turn>'
print_info: UNK token        = 3 '<unk>'
print_info: PAD token        = 0 '<pad>'
print_info: LF token         = 248 '<0x0A>'
print_info: EOG token        = 106 '<end_of_turn>'
print_info: max token length = 48
load_tensors: loading model tensors, this can take a while... (mmap = true)
[DEBUG] <Mistral-Small-24B-Instruct-2501> Health check error on http://localhost:5802/health, Get "http://localhost:5802/health": context deadline exceeded (Client.Timeout exceeded while awaiting headers) (normal during startup)
srv  log_server_r: request: GET /health 127.0.0.1 503
srv  log_server_r: request: GET /health 127.0.0.1 503
[DEBUG] <EtherealAurora-12B-v2> Health check error on http://localhost:5800/health, status code: 503 (normal during startup)
[DEBUG] <gemma-3-27b-it> Health check error on http://localhost:5808/health, status code: 503 (normal during startup)
[DEBUG] <Mistral-Small-24B-Instruct-2501> Health check error on http://localhost:5802/health, Get "http://localhost:5802/health": context deadline exceeded (Client.Timeout exceeded while awaiting headers) (normal during startup)
srv  log_server_r: request: GET /health 127.0.0.1 503
[DEBUG] <EtherealAurora-12B-v2> Health check error on http://localhost:5800/health, status code: 503 (normal during startup)
srv  log_server_r: request: GET /health 127.0.0.1 503
[DEBUG] <gemma-3-27b-it> Health check error on http://localhost:5808/health, status code: 503 (normal during startup)
[DEBUG] <Mistral-Small-24B-Instruct-2501> Health check error on http://localhost:5802/health, Get "http://localhost:5802/health": context deadline exceeded (Client.Timeout exceeded while awaiting headers) (normal during startup)
srv  log_server_r: request: GET /health 127.0.0.1 503
srv  log_server_r: request: GET /health 127.0.0.1 503
[DEBUG] <EtherealAurora-12B-v2> Health check error on http://localhost:5800/health, status code: 503 (normal during startup)
[DEBUG] <gemma-3-27b-it> Health check error on http://localhost:5808/health, status code: 503 (normal during startup)
[DEBUG] <Mistral-Small-24B-Instruct-2501> Health check error on http://localhost:5802/health, Get "http://localhost:5802/health": context deadline exceeded (Client.Timeout exceeded while awaiting headers) (normal during startup)
[DEBUG] <gemma-3-27b-it> Health check error on http://localhost:5808/health, status code: 503 (normal during startup)
srv  log_server_r: request: GET /health 127.0.0.1 503
[DEBUG] <EtherealAurora-12B-v2> Health check error on http://localhost:5800/health, status code: 503 (normal during startup)
srv  log_server_r: request: GET /health 127.0.0.1 503
[DEBUG] <Mistral-Small-24B-Instruct-2501> Health check error on http://localhost:5802/health, Get "http://localhost:5802/health": context deadline exceeded (Client.Timeout exceeded while awaiting headers) (normal during startup)

[mostlygeek]

@Garaymie thanks for reporting this. I fixed it in #277. It was a small fix but tricky to get testing right.