{
 "case": "feat-multimodal",
 "machine": "windows-2025-nocuda",
 "run": "ci-2",
 "args": [
  "--model",
  "models\\SmolVLM-256M-Instruct-Q8_0.gguf",
  "--mmproj",
  "models\\mmproj-SmolVLM-256M-Instruct-Q8_0.gguf",
  "--whispermodel",
  "models\\ggml-tiny.en.bin",
  "--embeddingsmodel",
  "models\\all-MiniLM-L6-v2-Q8_0.gguf",
  "--contextsize",
  "4096"
 ],
 "started": "2026-11-13T13:15:05+0000",
 "start": {
  "ready": true,
  "seconds": 13.61,
  "model": {
   "result": "koboldcpp/SmolVLM-256M-Instruct-Q8_0"
  }
 },
 "mm": {
  "version": {
   "result": "KoboldCpp",
   "version": "1.124.2",
   "protected": false,
   "llm": true,
   "txt2img": false,
   "vision": true,
   "audio": false,
   "transcribe": true,
   "multiplayer": false,
   "websearch": false,
   "tts": false,
   "embeddings": true,
   "music": false,
   "musicllm": false,
   "musicfamily": "none",
   "savedata": false,
   "admin": 0,
   "router": false,
   "guidance": false,
   "jinja": true,
   "mcp": false,
   "parallelrequests": 1,
   "defaults_modified": false
  },
  "vision": {
   "status": 200,
   "s": 5.95,
   "text": " \\red\\quad\\blue\\quad",
   "prompt_tokens": 159
  },
  "whisper": {
   "status": 200,
   "s": 1.9,
   "text": " And so my fellow Americans ask not what your country can do for you, ask what you can do for your country."
  },
  "embeddings": {
   "status": 200,
   "s": 0.062,
   "dim": 384,
   "cos_cat_kitten": 0.659,
   "cos_cat_stock": 0.077
  }
 },
 "peak_rss_mb": 745.5,
 "log_lines": [
  "Unable to detect VRAM.",
  "Auto Selected Default Backend (flag=0)",
  "Unable to detect VRAM.",
  "Unable to determine GPU Memory",
  "Detected Available RAM: 13601 MB",
  "Initializing dynamic library: koboldcpp_default.dll",
  "Namespace(admin=False, admindir='', adminpassword=<redacted> adminunloadtimeout=0, agent=False, agent_api_key=<redacted> agent_base_url=None, agentload='', allow_config_onready=False, analyze='', autofit=False, autofitpadding=1024, autoswapmode=False, autoswapthreshold=256, baseconfig='', batchsize=512, benchmark=None, blasthreads=0, chatcompletionsadapter='AutoGuess', cli=False, config=None, contextsize=4096, debugmode=0, defaultgenamt=2048, device='', downloaddir='', draftamount=4, draftgpulayers=999, draftgpusplit=None, draftmodel='', embeddingsgpu=False, embeddingsmaxctx=4096, embeddingsmodel='models\\\\all-MiniLM-L6-v2-Q8_0.gguf', enableguidance=False, exportconfig='', exporttemplate='', failsafe=False, ffncpu=0, flashattention=False, forceversion=False, foreground=False, gendefaults='', gendefaultsoverwrite=False, genlimit=0, gpulayers=0, highpriority=False, hordeconfig=None, hordegenlen=0, hordekey='', hordemaxctx=0, hordemodelname='', hordeworkername='', host='', ignoremissing=False, jinja=False, jinja_kwargs='', jinjatemplate='', jinjathink='default', launch=False, lazy_mode='auto', lora=None, loramult=1.0, lowvram=False, maingpu=-1, maxrequestsize=32, mcpfile='', mmproj='models\\\\mmproj-SmolVLM-256M-Instruct-Q8_0.gguf', mmprojcpu=False, model=['models\\\\SmolVLM-256M-Instruct-Q8_0.gguf'], model_param='models\\\\SmolVLM-256M-Instruct-Q8_0.gguf', moe_cache_mib=0, moecpu=0, moeexperts=-1, multiplayer=False, multiuser=None, musicdiffusion='', musicembeddings='', musicllm='', musiclowvram=False, musicvae='', noavx2=False, noblas=False, nobostoken=<redacted> nocertify=False, nofastforward=False, noflashattention=False, nommap=False, nommq=False, nomodel=False, nopipelineparallel=False, noshift=False, noswa=False, onready='', overridekv='', overridenativecontext=0, overridetensors='', parallelrequests=1, paralleltotalcontext=0, password=<redacted> pipelineparallel=False, port=49248, port_param=49248, preloadstory='', prompt='', proxy_port=None, quantkv='f16', queuelimit=10, quiet=False, ratelimit=0, reasoningeffort='default', remotetunnel=False, reqtimeout=600, ropeconfig=[0.0, 10000.0], routermode=False, rpccache='', rpcdevice='', rpchost='0.0.0.0', rpcmode='disabled', rpcport=5551, rpctargets='', run_bundled_agent=False, savedatafile='', sdaudiovae='', sdclamped=0, sdclampedsoft=0, sdclip1='', sdclip2='', sdclipdevice=-2, sdclipgpu=False, sdconfig=None, sdconvdirect='off', sdflashattention=False, sdgendefaults=False, sdllm='', sdlora=[], sdloramult=[1.0], sdmaingpu=None, sdmodel='', sdnotile=False, sdoffloadcpu=False, sdphotomaker='', sdquant=0, sdt5xxl=None, sdthreads=0, sdtiledvae=512, sdupscaler='', sdvae='', sdvaeauto=False, sdvaecpu=False, sdvaedevice=-1, sdvramlimit=0, showgui=False, singleinstance=False, skiplauncher=True, smartcache=0, smartcontext=False, splitmode='layer', ssl=None, swapadding=0, tensor_split=None, testmemory=False, threads=2, ttsdir='', ttsgpu=False, ttsmaxlen=4096, ttsmodel='', ttsthreads=0, ttswavtokenizer='', ubatchsize=-1, unpack='', usecpu=False, usecuda=None, usedirectio=False, usemlock=False, usemmap=False, usemtp=False, useswa=False, usevulkan=None, version=False, visionmaxres=1024, visionmaxtokens=-1, visionmintokens=-1, websearch=False, whispermodel='models\\\\ggml-tiny.en.bin')",
  "print_info: file size   = 165.24 MiB (8.51 BPW)",
  "print_info: n_ctx_train           = 8192",
  "print_info: n_ctx_orig_yarn       = 8192",
  "done_getting_tensors: tensor 'token_embd.weight' (q8_0) (and 272 others) moved from CPU_REPACK, using CPU instead",
  "load_tensors:          CPU model buffer size =   165.24 MiB",
  "llama_context: n_ctx                 = 4352",
  "llama_context: n_ctx_seq             = 4352",
  "llama_context: flash_attn            = enabled",
  "llama_context: n_ctx_seq (4352) < n_ctx_train (8192) -- the full capacity of the model will not be utilized",
  "llama_kv_cache:        CPU KV buffer size =    95.62 MiB",
  "sched_reserve:        CPU compute buffer size =   107.27 MiB",
  "print_info: file size   = 23.13 MiB (8.60 BPW)",
  "print_info: n_ctx_train           = 512",
  "print_info: n_ctx_orig_yarn       = 512",
  "done_getting_tensors: tensor 'token_embd.weight' (q8_0) (and 100 others) moved from CPU_REPACK, using CPU instead",
  "load_tensors:          CPU model buffer size =    23.13 MiB",
  "llama_context: n_ctx                 = 512",
  "llama_context: n_ctx_seq             = 512",
  "llama_context: flash_attn            = enabled",
  "Load Text Model OK: True",
  "Chat template heuristics failed to identify chat completions format. Jinja will be used.",
  "Starting Kobold API on port 49248 at http://localhost:49248/api/",
  "Please connect to custom endpoint at http://localhost:49248",
  "Threadpool set to 2 threads and 2 blasthreads...",
  "reserve_compute_meta:        CPU compute buffer size =    21.00 MiB"
 ],
 "finished": "2026-11-13T13:15:27+0000"
}