huangfu/DronePlanning
2
1
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

AI重构

Browse Source
This commit is contained in:
huangfu
2026-02-20 23:04:09 +08:00
parent 43d69d5a99
commit 5d8412bcb6
244 changed files with 9304 additions and 1463 deletions
Download Patch File Download Diff File Expand all files Collapse all files

145
logs/inference_model.log
View File

@@ -11,7 +11,7 @@ main: binding port with default address family
main: HTTP server is listening, hostname: 0.0.0.0, port: 8081, http threads: 15
main: loading model
srv load_model: loading model '/home/huangfukk/models/gguf/Qwen3/Qwen3-4B/Qwen3-4B-Q5_K_M.gguf'
llama_model_load_from_file_impl: using device CUDA0 (NVIDIA GeForce RTX 4060 Ti) - 2789 MiB free
llama_model_load_from_file_impl: using device CUDA0 (NVIDIA GeForce RTX 4060 Ti) - 15225 MiB free
llama_model_loader: loaded meta data with 28 key-value pairs and 398 tensors from /home/huangfukk/models/gguf/Qwen3/Qwen3-4B/Qwen3-4B-Q5_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 = qwen3
@@ -112,6 +112,7 @@ print_info: EOG token = 151663 '<|repo_name|>'
print_info: EOG token = 151664 '<|file_sep|>'
print_info: max token length = 256
load_tensors: loading model tensors, this can take a while... (mmap = true)
srv log_server_r: request: GET /health 127.0.0.1 503
load_tensors: offloading 36 repeating layers to GPU
load_tensors: offloaded 36/37 layers to GPU
load_tensors: CUDA0 model buffer size = 2445.68 MiB
@@ -119,8 +120,8 @@ load_tensors: CPU_Mapped model buffer size = 304.29 MiB
..........................................................................................
llama_context: constructing llama_context
llama_context: n_seq_max = 1
llama_context: n_ctx = 8192
llama_context: n_ctx_per_seq = 8192
llama_context: n_ctx = 16384
llama_context: n_ctx_per_seq = 16384
llama_context: n_batch = 2048
llama_context: n_ubatch = 512
llama_context: causal_attn = 1
@@ -128,12 +129,12 @@ llama_context: flash_attn = 0
llama_context: kv_unified = false
llama_context: freq_base = 1000000.0
llama_context: freq_scale = 1
llama_context: n_ctx_per_seq (8192) < n_ctx_train (40960) -- the full capacity of the model will not be utilized
llama_context: n_ctx_per_seq (16384) < n_ctx_train (40960) -- the full capacity of the model will not be utilized
llama_context: CPU output buffer size = 0.58 MiB
llama_kv_cache_unified: CUDA0 KV buffer size = 1152.00 MiB
llama_kv_cache_unified: size = 1152.00 MiB ( 8192 cells, 36 layers, 1/1 seqs), K (f16): 576.00 MiB, V (f16): 576.00 MiB
llama_context: CUDA0 compute buffer size = 606.03 MiB
llama_context: CUDA_Host compute buffer size = 25.01 MiB
llama_kv_cache_unified: CUDA0 KV buffer size = 2304.00 MiB
llama_kv_cache_unified: size = 2304.00 MiB ( 16384 cells, 36 layers, 1/1 seqs), K (f16): 1152.00 MiB, V (f16): 1152.00 MiB
llama_context: CUDA0 compute buffer size = 1086.01 MiB
llama_context: CUDA_Host compute buffer size = 41.01 MiB
llama_context: graph nodes = 1410
llama_context: graph splits = 4 (with bs=512), 3 (with bs=1)
common_init_from_params: added <|endoftext|> logit bias = -inf
@@ -141,10 +142,10 @@ common_init_from_params: added <|im_end|> logit bias = -inf
common_init_from_params: added <|fim_pad|> logit bias = -inf
common_init_from_params: added <|repo_name|> logit bias = -inf
common_init_from_params: added <|file_sep|> logit bias = -inf
common_init_from_params: setting dry_penalty_last_n to ctx_size = 8192
common_init_from_params: setting dry_penalty_last_n to ctx_size = 16384
common_init_from_params: warming up the model with an empty run - please wait ... (--no-warmup to disable)
srv init: initializing slots, n_slots = 1
slot init: id 0 | task -1 | new slot n_ctx_slot = 8192
slot init: id 0 | task -1 | new slot n_ctx_slot = 16384
main: model loaded
main: chat template, chat_template: {%- if tools %}
{{- '<|im_start|>system\n' }}
@@ -242,8 +243,122 @@ How are you?<|im_end|>
'
main: server is listening on http://0.0.0.0:8081 - starting the main loop
srv update_slots: all slots are idle
srv operator(): operator(): cleaning up before exit...
ed second interrupt, terminating immediately.
srv operator(): operator(): cleaning up before exit...
srv log_server_r: request: GET /health 127.0.0.1 200
srv operator(): operator(): cleaning up before exit...
srv log_server_r: request: GET /v1/models 127.0.0.1 200
srv params_from_: Chat format: Content-only
slot launch_slot_: id 0 | task 0 | processing task
slot update_slots: id 0 | task 0 | new prompt, n_ctx_slot = 16384, n_keep = 0, n_prompt_tokens = 1569
slot update_slots: id 0 | task 0 | kv cache rm [0, end)
slot update_slots: id 0 | task 0 | prompt processing progress, n_past = 1569, n_tokens = 1569, progress = 1.000000
slot update_slots: id 0 | task 0 | prompt done, n_past = 1569, n_tokens = 1569
slot release: id 0 | task 0 | stop processing: n_past = 1578, truncated = 0
slot print_timing: id 0 | task 0 |
prompt eval time = 1012.37 ms / 1569 tokens ( 0.65 ms per token, 1549.82 tokens per second)
eval time = 258.11 ms / 10 tokens ( 25.81 ms per token, 38.74 tokens per second)
total time = 1270.48 ms / 1579 tokens
srv update_slots: all slots are idle
srv log_server_r: request: POST /v1/chat/completions 127.0.0.1 200
srv params_from_: Chat format: Content-only
slot launch_slot_: id 0 | task 11 | processing task
slot update_slots: id 0 | task 11 | new prompt, n_ctx_slot = 16384, n_keep = 0, n_prompt_tokens = 1611
slot update_slots: id 0 | task 11 | kv cache rm [5, end)
slot update_slots: id 0 | task 11 | prompt processing progress, n_past = 1611, n_tokens = 1606, progress = 0.996896
slot update_slots: id 0 | task 11 | prompt done, n_past = 1611, n_tokens = 1606
slot release: id 0 | task 11 | stop processing: n_past = 1636, truncated = 0
slot print_timing: id 0 | task 11 |
prompt eval time = 613.51 ms / 1606 tokens ( 0.38 ms per token, 2617.71 tokens per second)
eval time = 653.35 ms / 26 tokens ( 25.13 ms per token, 39.80 tokens per second)
total time = 1266.86 ms / 1632 tokens
srv update_slots: all slots are idle
srv log_server_r: request: POST /v1/chat/completions 127.0.0.1 200
srv params_from_: Chat format: Content-only
slot launch_slot_: id 0 | task 38 | processing task
slot update_slots: id 0 | task 38 | new prompt, n_ctx_slot = 16384, n_keep = 0, n_prompt_tokens = 1574
slot update_slots: id 0 | task 38 | kv cache rm [5, end)
slot update_slots: id 0 | task 38 | prompt processing progress, n_past = 1574, n_tokens = 1569, progress = 0.996823
slot update_slots: id 0 | task 38 | prompt done, n_past = 1574, n_tokens = 1569
slot release: id 0 | task 38 | stop processing: n_past = 1584, truncated = 0
slot print_timing: id 0 | task 38 |
prompt eval time = 768.98 ms / 1569 tokens ( 0.49 ms per token, 2040.36 tokens per second)
eval time = 254.75 ms / 11 tokens ( 23.16 ms per token, 43.18 tokens per second)
total time = 1023.74 ms / 1580 tokens
srv update_slots: all slots are idle
srv log_server_r: request: POST /v1/chat/completions 127.0.0.1 200
srv params_from_: Chat format: Content-only
slot launch_slot_: id 0 | task 50 | processing task
slot update_slots: id 0 | task 50 | new prompt, n_ctx_slot = 16384, n_keep = 0, n_prompt_tokens = 4929
slot update_slots: id 0 | task 50 | kv cache rm [3, end)
slot update_slots: id 0 | task 50 | prompt processing progress, n_past = 2051, n_tokens = 2048, progress = 0.415500
slot update_slots: id 0 | task 50 | kv cache rm [2051, end)
slot update_slots: id 0 | task 50 | prompt processing progress, n_past = 4099, n_tokens = 2048, progress = 0.831000
slot update_slots: id 0 | task 50 | kv cache rm [4099, end)
slot update_slots: id 0 | task 50 | prompt processing progress, n_past = 4929, n_tokens = 830, progress = 0.999391
slot update_slots: id 0 | task 50 | prompt done, n_past = 4929, n_tokens = 830
slot release: id 0 | task 50 | stop processing: n_past = 5254, truncated = 0
slot print_timing: id 0 | task 50 |
prompt eval time = 2682.34 ms / 4926 tokens ( 0.54 ms per token, 1836.46 tokens per second)
eval time = 9502.10 ms / 326 tokens ( 29.15 ms per token, 34.31 tokens per second)
total time = 12184.44 ms / 5252 tokens
srv update_slots: all slots are idle
srv log_server_r: request: POST /v1/chat/completions 127.0.0.1 200
srv params_from_: Chat format: Content-only
slot launch_slot_: id 0 | task 379 | processing task
slot update_slots: id 0 | task 379 | new prompt, n_ctx_slot = 16384, n_keep = 0, n_prompt_tokens = 1573
slot update_slots: id 0 | task 379 | kv cache rm [3, end)
slot update_slots: id 0 | task 379 | prompt processing progress, n_past = 1573, n_tokens = 1570, progress = 0.998093
slot update_slots: id 0 | task 379 | prompt done, n_past = 1573, n_tokens = 1570
slot release: id 0 | task 379 | stop processing: n_past = 1583, truncated = 0
slot print_timing: id 0 | task 379 |
prompt eval time = 862.21 ms / 1570 tokens ( 0.55 ms per token, 1820.91 tokens per second)
eval time = 284.82 ms / 11 tokens ( 25.89 ms per token, 38.62 tokens per second)
total time = 1147.03 ms / 1581 tokens
srv update_slots: all slots are idle
srv log_server_r: request: POST /v1/chat/completions 127.0.0.1 200
srv params_from_: Chat format: Content-only
slot launch_slot_: id 0 | task 391 | processing task
slot update_slots: id 0 | task 391 | new prompt, n_ctx_slot = 16384, n_keep = 0, n_prompt_tokens = 7208
slot update_slots: id 0 | task 391 | kv cache rm [3, end)
slot update_slots: id 0 | task 391 | prompt processing progress, n_past = 2051, n_tokens = 2048, progress = 0.284129
slot update_slots: id 0 | task 391 | kv cache rm [2051, end)
slot update_slots: id 0 | task 391 | prompt processing progress, n_past = 4099, n_tokens = 2048, progress = 0.568258
slot update_slots: id 0 | task 391 | kv cache rm [4099, end)
slot update_slots: id 0 | task 391 | prompt processing progress, n_past = 6147, n_tokens = 2048, progress = 0.852386
slot update_slots: id 0 | task 391 | kv cache rm [6147, end)
slot update_slots: id 0 | task 391 | prompt processing progress, n_past = 7208, n_tokens = 1061, progress = 0.999584
slot update_slots: id 0 | task 391 | prompt done, n_past = 7208, n_tokens = 1061
slot release: id 0 | task 391 | stop processing: n_past = 7577, truncated = 0
slot print_timing: id 0 | task 391 |
prompt eval time = 4810.49 ms / 7205 tokens ( 0.67 ms per token, 1497.77 tokens per second)
eval time = 11443.72 ms / 370 tokens ( 30.93 ms per token, 32.33 tokens per second)
total time = 16254.21 ms / 7575 tokens
srv update_slots: all slots are idle
srv log_server_r: request: POST /v1/chat/completions 127.0.0.1 200
srv params_from_: Chat format: Content-only
slot launch_slot_: id 0 | task 765 | processing task
slot update_slots: id 0 | task 765 | new prompt, n_ctx_slot = 16384, n_keep = 0, n_prompt_tokens = 1572
slot update_slots: id 0 | task 765 | kv cache rm [3, end)
slot update_slots: id 0 | task 765 | prompt processing progress, n_past = 1572, n_tokens = 1569, progress = 0.998092
slot update_slots: id 0 | task 765 | prompt done, n_past = 1572, n_tokens = 1569
slot release: id 0 | task 765 | stop processing: n_past = 1582, truncated = 0
slot print_timing: id 0 | task 765 |
prompt eval time = 826.51 ms / 1569 tokens ( 0.53 ms per token, 1898.34 tokens per second)
eval time = 227.62 ms / 11 tokens ( 20.69 ms per token, 48.33 tokens per second)
total time = 1054.13 ms / 1580 tokens
srv update_slots: all slots are idle
srv log_server_r: request: POST /v1/chat/completions 127.0.0.1 200
srv params_from_: Chat format: Content-only
slot launch_slot_: id 0 | task 777 | processing task
slot update_slots: id 0 | task 777 | new prompt, n_ctx_slot = 16384, n_keep = 0, n_prompt_tokens = 4927
slot update_slots: id 0 | task 777 | kv cache rm [3, end)
slot update_slots: id 0 | task 777 | prompt processing progress, n_past = 2051, n_tokens = 2048, progress = 0.415669
slot update_slots: id 0 | task 777 | kv cache rm [2051, end)
slot update_slots: id 0 | task 777 | prompt processing progress, n_past = 4099, n_tokens = 2048, progress = 0.831338
slot update_slots: id 0 | task 777 | kv cache rm [4099, end)
slot update_slots: id 0 | task 777 | prompt processing progress, n_past = 4927, n_tokens = 828, progress = 0.999391
slot update_slots: id 0 | task 777 | prompt done, n_past = 4927, n_tokens = 828
slot release: id 0 | task 777 | stop processing: n_past = 5246, truncated = 0
slot print_timing: id 0 | task 777 |
prompt eval time = 2382.24 ms / 4924 tokens ( 0.48 ms per token, 2066.96 tokens per second)
eval time = 8172.57 ms / 320 tokens ( 25.54 ms per token, 39.16 tokens per second)
total time = 10554.82 ms / 5244 tokens
srv update_slots: all slots are idle
srv log_server_r: request: POST /v1/chat/completions 127.0.0.1 200