.\llama-server.exe -c 85000 -n 16384 -m C:\Temp\Qwen3.6-27B-UD-IQ3_XXS.gguf -ngl 99 -fa on --host 127.0.0.1 -a qwen3.5-27b --port 8080 --jinja -ctk q8_0 -ctv q8_0 --temperature 0.6 --top-p 0.95 --top-k 20 --min-p 0.0 --no-webui -np 1 --spec-type ngram-mod --spec-ngram-size-n 24 --draft-min 12 --draft-max 48 --chat-template-kwargs '{\"preserve_thinking\": true}'
ggml_cuda_init: found 1 CUDA devices (Total VRAM: 16375 MiB):
  Device 0: NVIDIA GeForce RTX 4070 Ti SUPER, compute capability 8.9, VMM: yes, VRAM: 16375 MiB
build_info: b8917-c3ef7631c
system_info: n_threads = 8 (n_threads_batch = 8) / 16 | CUDA : ARCHS = 890 | USE_GRAPHS = 1 | PEER_MAX_BATCH_SIZE = 128 | CPU : SSE3 = 1 | SSSE3 = 1 | AVX = 1 | AVX2 = 1 | F16C = 1 | FMA = 1 | AVX512 = 1 | LLAMAFILE = 1 | OPENMP = 1 | REPACK = 1 |
init: using 15 threads for HTTP server
Web UI is disabled
start: binding port with default address family
main: loading model
srv    load_model: loading model 'C:\Temp\Qwen3.6-27B-UD-IQ3_XXS.gguf'
common_init_result: fitting params to device memory, for bugs during this step try to reproduce them with -fit off, or provide --verbose logs if the bug only occurs with -fit on
common_params_fit_impl: getting device memory data for initial parameters:
common_memory_breakdown_print: | memory breakdown [MiB]        | total    free     self   model   context   compute       unaccounted |
common_memory_breakdown_print: |   - CUDA0 (RTX 4070 Ti SUPER) | 16375 = 14935 + (14382 = 10907 +    2980 +     495) + 17592186031473 |
common_memory_breakdown_print: |   - Host                      |                    707 =   520 +       0 +     186                   |
common_params_fit_impl: projected to use 14382 MiB of device memory vs. 14935 MiB of free device memory
common_params_fit_impl: cannot meet free memory target of 1024 MiB, need to reduce device memory by 471 MiB
common_params_fit_impl: context size set by user to 85000 -> no change
common_fit_params: failed to fit params to free device memory: n_gpu_layers already set by user to 99, abort
�[0mcommon_fit_params: fitting params to free memory took 0.40 seconds
llama_model_load_from_file_impl: using device CUDA0 (NVIDIA GeForce RTX 4070 Ti SUPER) (0000:01:00.0) - 15085 MiB free
llama_model_loader: loaded meta data with 51 key-value pairs and 851 tensors from C:\Temp\Qwen3.6-27B-UD-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              = qwen35
llama_model_loader: - kv   1:                               general.type str              = model
llama_model_loader: - kv   2:                     general.sampling.top_k i32              = 20
llama_model_loader: - kv   3:                     general.sampling.top_p f32              = 0.950000
llama_model_loader: - kv   4:                      general.sampling.temp f32              = 1.000000
llama_model_loader: - kv   5:                               general.name str              = Qwen3.6-27B
llama_model_loader: - kv   6:                           general.basename str              = Qwen3.6-27B
llama_model_loader: - kv   7:                       general.quantized_by str              = Unsloth
llama_model_loader: - kv   8:                         general.size_label str              = 27B
llama_model_loader: - kv   9:                            general.license str              = apache-2.0
llama_model_loader: - kv  10:                       general.license.link str              = https://huggingface.co/Qwen/Qwen3.6-2...
llama_model_loader: - kv  11:                           general.repo_url str              = https://huggingface.co/unsloth
llama_model_loader: - kv  12:                   general.base_model.count u32              = 1
llama_model_loader: - kv  13:                  general.base_model.0.name str              = Qwen3.6 27B
llama_model_loader: - kv  14:          general.base_model.0.organization str              = Qwen
llama_model_loader: - kv  15:              general.base_model.0.repo_url str              = https://huggingface.co/Qwen/Qwen3.6-27B
llama_model_loader: - kv  16:                               general.tags arr[str,2]       = ["unsloth", "image-text-to-text"]
llama_model_loader: - kv  17:                         qwen35.block_count u32              = 64
llama_model_loader: - kv  18:                      qwen35.context_length u32              = 262144
llama_model_loader: - kv  19:                    qwen35.embedding_length u32              = 5120
llama_model_loader: - kv  20:                 qwen35.feed_forward_length u32              = 17408
llama_model_loader: - kv  21:                qwen35.attention.head_count u32              = 24
llama_model_loader: - kv  22:             qwen35.attention.head_count_kv u32              = 4
llama_model_loader: - kv  23:             qwen35.rope.dimension_sections arr[i32,4]       = [11, 11, 10, 0]
llama_model_loader: - kv  24:                      qwen35.rope.freq_base f32              = 10000000.000000
llama_model_loader: - kv  25:    qwen35.attention.layer_norm_rms_epsilon f32              = 0.000001
llama_model_loader: - kv  26:                qwen35.attention.key_length u32              = 256
llama_model_loader: - kv  27:              qwen35.attention.value_length u32              = 256
llama_model_loader: - kv  28:                     qwen35.ssm.conv_kernel u32              = 4
llama_model_loader: - kv  29:                      qwen35.ssm.state_size u32              = 128
llama_model_loader: - kv  30:                     qwen35.ssm.group_count u32              = 16
llama_model_loader: - kv  31:                  qwen35.ssm.time_step_rank u32              = 48
llama_model_loader: - kv  32:                      qwen35.ssm.inner_size u32              = 6144
llama_model_loader: - kv  33:             qwen35.full_attention_interval u32              = 4
llama_model_loader: - kv  34:                qwen35.rope.dimension_count u32              = 64
llama_model_loader: - kv  35:                       tokenizer.ggml.model str              = gpt2
llama_model_loader: - kv  36:                         tokenizer.ggml.pre str              = qwen35
llama_model_loader: - kv  37:                      tokenizer.ggml.tokens arr[str,248320]  = ["!", "\"", "#", "$", "%", "&", "'", ...
llama_model_loader: - kv  38:                  tokenizer.ggml.token_type arr[i32,248320]  = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
llama_model_loader: - kv  39:                      tokenizer.ggml.merges arr[str,247587]  = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",...
llama_model_loader: - kv  40:                tokenizer.ggml.eos_token_id u32              = 248046
llama_model_loader: - kv  41:            tokenizer.ggml.padding_token_id u32              = 248055
llama_model_loader: - kv  42:                tokenizer.ggml.bos_token_id u32              = 248044
llama_model_loader: - kv  43:               tokenizer.ggml.add_bos_token bool             = false
llama_model_loader: - kv  44:                    tokenizer.chat_template str              = {%- set image_count = namespace(value...
llama_model_loader: - kv  45:               general.quantization_version u32              = 2
llama_model_loader: - kv  46:                          general.file_type u32              = 23
llama_model_loader: - kv  47:                      quantize.imatrix.file str              = Qwen3.6-27B-GGUF/imatrix_unsloth.gguf
llama_model_loader: - kv  48:                   quantize.imatrix.dataset str              = unsloth_calibration_Qwen3.6-27B.txt
llama_model_loader: - kv  49:             quantize.imatrix.entries_count u32              = 496
llama_model_loader: - kv  50:              quantize.imatrix.chunks_count u32              = 76
llama_model_loader: - type  f32:  449 tensors
llama_model_loader: - type q3_K:    1 tensors
llama_model_loader: - type q4_K:   48 tensors
llama_model_loader: - type q5_K:    1 tensors
llama_model_loader: - type q6_K:   48 tensors
llama_model_loader: - type iq2_xs:    2 tensors
llama_model_loader: - type iq3_xxs:  254 tensors
llama_model_loader: - type iq3_s:   19 tensors
llama_model_loader: - type iq2_s:   24 tensors
llama_model_loader: - type iq4_xs:    5 tensors
print_info: file format = GGUF V3 (latest)
print_info: file type   = IQ3_XXS - 3.0625 bpw
print_info: file size   = 11.16 GiB (3.56 BPW)
load: 0 unused tokens
load: printing all EOG tokens:
load:   - 248044 ('<|endoftext|>')
load:   - 248046 ('<|im_end|>')
load:   - 248063 ('<|fim_pad|>')
load:   - 248064 ('<|repo_name|>')
load:   - 248065 ('<|file_sep|>')
load: special tokens cache size = 33
load: token to piece cache size = 1.7581 MB
print_info: arch                  = qwen35
print_info: vocab_only            = 0
print_info: no_alloc              = 0
print_info: n_ctx_train           = 262144
print_info: n_embd                = 5120
print_info: n_embd_inp            = 5120
print_info: n_layer               = 64
print_info: n_head                = 24
print_info: n_head_kv             = 4
print_info: n_rot                 = 64
print_info: n_swa                 = 0
print_info: is_swa_any            = 0
print_info: n_embd_head_k         = 256
print_info: n_embd_head_v         = 256
print_info: n_gqa                 = 6
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-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          = 0.0e+00
print_info: n_ff                  = 17408
print_info: n_expert              = 0
print_info: n_expert_used         = 0
print_info: n_expert_groups       = 0
print_info: n_group_used          = 0
print_info: causal attn           = 1
print_info: pooling type          = -1
print_info: rope type             = 40
print_info: rope scaling          = linear
print_info: freq_base_train       = 10000000.0
print_info: freq_scale_train      = 1
print_info: n_ctx_orig_yarn       = 262144
print_info: rope_yarn_log_mul     = 0.0000
print_info: rope_finetuned        = unknown
print_info: mrope sections        = [11, 11, 10, 0]
print_info: ssm_d_conv            = 4
print_info: ssm_d_inner           = 6144
print_info: ssm_d_state           = 128
print_info: ssm_dt_rank           = 48
print_info: ssm_n_group           = 16
print_info: ssm_dt_b_c_rms        = 0
print_info: model type            = 27B
print_info: model params          = 26.90 B
print_info: general.name          = Qwen3.6-27B
print_info: vocab type            = BPE
print_info: n_vocab               = 248320
print_info: n_merges              = 247587
print_info: BOS token             = 248044 '<|endoftext|>'
print_info: EOS token             = 248046 '<|im_end|>'
print_info: EOT token             = 248046 '<|im_end|>'
print_info: PAD token             = 248055 '<|vision_pad|>'
print_info: LF token              = 198 'Ċ'
print_info: FIM PRE token         = 248060 '<|fim_prefix|>'
print_info: FIM SUF token         = 248062 '<|fim_suffix|>'
print_info: FIM MID token         = 248061 '<|fim_middle|>'
print_info: FIM PAD token         = 248063 '<|fim_pad|>'
print_info: FIM REP token         = 248064 '<|repo_name|>'
print_info: FIM SEP token         = 248065 '<|file_sep|>'
print_info: EOG token             = 248044 '<|endoftext|>'
print_info: EOG token             = 248046 '<|im_end|>'
print_info: EOG token             = 248063 '<|fim_pad|>'
print_info: EOG token             = 248064 '<|repo_name|>'
print_info: EOG token             = 248065 '<|file_sep|>'
print_info: max token length      = 256
load_tensors: loading model tensors, this can take a while... (mmap = true, direct_io = false)
load_tensors: offloading output layer to GPU
load_tensors: offloading 63 repeating layers to GPU
load_tensors: offloaded 65/65 layers to GPU
load_tensors:   CPU_Mapped model buffer size =   521.00 MiB
load_tensors:        CUDA0 model buffer size = 10907.64 MiB
...........................................................................................
common_init_result: added <|endoftext|> logit bias = -inf
common_init_result: added <|im_end|> logit bias = -inf
common_init_result: added <|fim_pad|> logit bias = -inf
common_init_result: added <|repo_name|> logit bias = -inf
common_init_result: added <|file_sep|> logit bias = -inf
llama_context: constructing llama_context
llama_context: n_seq_max     = 1
llama_context: n_ctx         = 85248
llama_context: n_ctx_seq     = 85248
llama_context: n_batch       = 2048
llama_context: n_ubatch      = 512
llama_context: causal_attn   = 1
llama_context: flash_attn    = enabled
llama_context: kv_unified    = false
llama_context: freq_base     = 10000000.0
llama_context: freq_scale    = 1
llama_context: n_ctx_seq (85248) < n_ctx_train (262144) -- the full capacity of the model will not be utilized
�[0mllama_context:  CUDA_Host  output buffer size =     0.95 MiB
llama_kv_cache:      CUDA0 KV buffer size =  2830.50 MiB
llama_kv_cache: size = 2830.50 MiB ( 85248 cells,  16 layers,  1/1 seqs), K (q8_0): 1415.25 MiB, V (q8_0): 1415.25 MiB
llama_kv_cache: attn_rot_k = 1, n_embd_head_k_all = 256
llama_kv_cache: attn_rot_v = 1, n_embd_head_k_all = 256
llama_memory_recurrent:      CUDA0 RS buffer size =   149.62 MiB
llama_memory_recurrent: size =  149.62 MiB (     1 cells,  64 layers,  1 seqs), R (f32):    5.62 MiB, S (f32):  144.00 MiB
sched_reserve: reserving ...
sched_reserve: resolving fused Gated Delta Net support:
sched_reserve: fused Gated Delta Net (autoregressive) enabled
sched_reserve: fused Gated Delta Net (chunked) enabled
sched_reserve:      CUDA0 compute buffer size =   495.00 MiB
sched_reserve:  CUDA_Host compute buffer size =   186.79 MiB
sched_reserve: graph nodes  = 3849
sched_reserve: graph splits = 2
sched_reserve: reserve took 16.73 ms, sched copies = 1
common_init_from_params: warming up the model with an empty run - please wait ... (--no-warmup to disable)
�[0msrv    load_model: initializing slots, n_slots = 1
common_context_can_seq_rm: the target context does not support partial sequence removal
�[0msrv    load_model: speculative decoding will use checkpoints
�[0mcommon_speculative_init: initialized ngram_mod with n=24, size=4194304 (16.000 MB)
slot   load_model: id  0 | task -1 | speculative decoding context initialized
slot   load_model: id  0 | task -1 | new slot, n_ctx = 85248
srv    load_model: prompt cache is enabled, size limit: 8192 MiB
�[0msrv    load_model: use `--cache-ram 0` to disable the prompt cache
�[0msrv    load_model: for more info see https://github.com/ggml-org/llama.cpp/pull/16391
�[0msrv          init: init: --cache-idle-slots requires --kv-unified, disabling
�[0minit: chat template, example_format: '<|im_start|>system
You are a helpful assistant<|im_end|>
<|im_start|>user
Hello<|im_end|>
<|im_start|>assistant
<think>

</think>

Hi there<|im_end|>
<|im_start|>user
How are you?<|im_end|>
<|im_start|>assistant
<think>
'
srv          init: init: chat template, thinking = 1
main: model loaded
main: server is listening on http://127.0.0.1:8080
main: starting the main loop...
srv  update_slots: all slots are idle
srv  log_server_r: done request: GET /api/version 127.0.0.1 200
srv  log_server_r: done request: GET /api/tags 127.0.0.1 200
srv  log_server_r: done request: POST /api/show 127.0.0.1 200
srv  params_from_: Chat format: peg-native
slot get_availabl: id  0 | task -1 | selected slot by LRU, t_last = -1
srv  get_availabl: updating prompt cache
�[0msrv          load:  - looking for better prompt, base f_keep = -1.000, sim = 0.000
�[0msrv        update:  - cache state: 0 prompts, 0.000 MiB (limits: 8192.000 MiB, 85248 tokens, 8589934592 est)
�[0msrv  get_availabl: prompt cache update took 0.01 ms
�[0mparse: error parsing grammar: unknown escape at \d]+ "." [\d]+) ("~" ([a-z] [a-z0-9_]* "." [a-z] [a-z0-9_]* "." [a-z_] [a-z0-9_.]* ".v" [\d]+ "." [\d]+))* ("~" tool-mcp-spec-server-m-add-block-arg-body-schema-type-1{8,8} "\b-" tool-mcp-spec-server-m-add-block-arg-body-schema-type-1{4,4} "\b-" tool-mcp-spec-server-m-add-block-arg-body-schema-type-1{4,4} "\b-" tool-mcp-spec-server-m-add-block-arg-body-schema-type-1{4,4} "\b-" tool-mcp-spec-server-m-add-block-arg-body-schema-type-1{12,12})?) "\"" space