{
 "case": "base-q2_k",
 "machine": "lab-lnx",
 "run": "run1-2026-11-13-local",
 "args": [
  "--model",
  "/home/<user>/lab/models/qwen2.5-0.5b-instruct-q2_k.gguf",
  "--contextsize",
  "4096"
 ],
 "started": "2026-11-13T12:39:19+0000",
 "start": {
  "ready": true,
  "seconds": 3.51,
  "model": {
   "result": "koboldcpp/qwen2.5-0.5b-instruct-q2_k"
  }
 },
 "gen": [
  {
   "status": 200,
   "wall_s": 5.978,
   "text_head": " looked at the lighthouse. The lighthouse was still lit, but the light was dim. She had been there for a long time, and it was time to check the machine. She wa",
   "prompt_tokens": 38,
   "gen_tokens": 128,
   "pp_tps": 121.02,
   "tg_tps": 22.65
  },
  {
   "status": 200,
   "wall_s": 5.455,
   "text_head": " looked at the lighthouse. She had never seen anything like it before. The lighthouse was a beacon of hope, a beacon of light that guided ships to safety. She h",
   "prompt_tokens": 38,
   "gen_tokens": 128,
   "pp_tps": 24.39,
   "tg_tps": 23.88
  },
  {
   "status": 200,
   "wall_s": 5.557,
   "text_head": " looked at the lighthouse. She had never seen anything like it before. The lighthouse was a beacon of hope, a beacon of light that guided ships to safety. She h",
   "prompt_tokens": 38,
   "gen_tokens": 128,
   "pp_tps": 19.61,
   "tg_tps": 23.47
  }
 ],
 "long": [
  {
   "status": 200,
   "wall_s": 15.864,
   "text_head": " The quick brown fox jumps over the lazy dog near the old stone bridge. The quick brown fox jumps over the lazy dog near the old stone bridge. The quick",
   "prompt_tokens": 1650,
   "gen_tokens": 32,
   "pp_tps": 116.52,
   "tg_tps": 19.55
  },
  {
   "status": 200,
   "wall_s": 1.704,
   "text_head": " The quick brown fox jumps over the lazy dog near the old stone bridge. The quick brown fox jumps over the lazy dog near the old stone bridge. The quick",
   "prompt_tokens": 1650,
   "gen_tokens": 32,
   "pp_tps": 19.61,
   "tg_tps": 20.06
  }
 ],
 "version": {
  "result": "KoboldCpp",
  "version": "1.124.2",
  "protected": false,
  "llm": true,
  "txt2img": false,
  "vision": false,
  "audio": false,
  "transcribe": false,
  "multiplayer": false,
  "websearch": false,
  "tts": false,
  "embeddings": false,
  "music": false,
  "musicllm": false,
  "musicfamily": "none",
  "savedata": false,
  "admin": 0,
  "router": false,
  "guidance": false,
  "jinja": false,
  "mcp": false,
  "parallelrequests": 1,
  "defaults_modified": false
 },
 "peak_rss_mb": 612.8,
 "log_lines": [
  "Unable to detect VRAM.",
  "Auto Selected Default Backend (flag=0)",
  "Unable to detect VRAM.",
  "Unable to determine GPU Memory",
  "Detected Available RAM: 15243 MB",
  "Initializing dynamic library: koboldcpp_default.so",
  "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='', 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='', mmprojcpu=False, model=['/home/<user>/lab/models/qwen2.5-0.5b-instruct-q2_k.gguf'], model_param='/home/<user>/lab/models/qwen2.5-0.5b-instruct-q2_k.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=47291, port_param=47291, 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='')",
  "print_info: file size   = 390.28 MiB (5.20 BPW)",
  "print_info: n_ctx_train           = 32768",
  "print_info: n_ctx_orig_yarn       = 32768",
  "done_getting_tensors: tensor 'token_embd.weight' (iq4_nl) (and 290 others) moved from CPU_REPACK, using CPU instead",
  "load_tensors:          CPU model buffer size =   390.28 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 (32768) -- the full capacity of the model will not be utilized",
  "llama_kv_cache:        CPU KV buffer size =    51.00 MiB",
  "sched_reserve:        CPU compute buffer size =   309.77 MiB",
  "Load Text Model OK: True",
  "Starting Kobold API on port 47291 at http://localhost:47291/api/",
  "Please connect to custom endpoint at http://localhost:47291",
  "Threadpool set to 2 threads and 2 blasthreads..."
 ],
 "finished": "2026-11-13T12:39:59+0000"
}