한번 찾아보니 가격 3배 이상, 토큰 3배 증가
사실상 9배 증가라는 소문이 있네
[링크 : https://wikidocs.net/blog/@jaehong/13849/]
근데 순식간에(?) geminu 3.0 flash 를 날리고 gemini 3.5 flash + 3.1 pro 를 같은 할당량으로 묶는건 너무 개매너 아니냐 -_-
[링크 : https://ai.google.dev/gemini-api/docs/pricing?hl=ko]
+
2026.05.29
느낌상 질문 하나당 한칸 줄어드는 느낌..
| python huggingface 저장경로 변경하기 (0) | 2026.05.24 |
|---|---|
| vllm 설치, 실행 실패 -> 사실상 포기 (0) | 2026.05.24 |
| openclaw agent 구성 관련 (0) | 2026.05.22 |
| mxfp4 (0) | 2026.05.22 |
| 텔레그램 봇 - 메시지 관련 (0) | 2026.05.22 |
| vllm 설치, 실행 실패 -> 사실상 포기 (0) | 2026.05.24 |
|---|---|
| antigravity gemini flash 할당량이 깃털 같구만? (0) | 2026.05.22 |
| mxfp4 (0) | 2026.05.22 |
| 텔레그램 봇 - 메시지 관련 (0) | 2026.05.22 |
| openclaw 설치 telegram 신기하네? (0) | 2026.05.22 |
microscale(mx) fp4
| antigravity gemini flash 할당량이 깃털 같구만? (0) | 2026.05.22 |
|---|---|
| openclaw agent 구성 관련 (0) | 2026.05.22 |
| 텔레그램 봇 - 메시지 관련 (0) | 2026.05.22 |
| openclaw 설치 telegram 신기하네? (0) | 2026.05.22 |
| wan2.2 + comfyui 시도 (0) | 2026.05.21 |
inlinebutton - 메시지 하단에 생기는 버튼
replybutton - 사용자 입력창 하단에 생기는 버튼
[링크 : https://naimjae.tistory.com/11]
메시지 서식(markdown 등)
[링크 : https://naimjae.tistory.com/8]
[링크 : https://docs.python-telegram-bot.org/en/stable/telegram.menubuttoncommands.html]
[링크 : https://docs.python-telegram-bot.org/en/stable/telegram.html]
+
2026.05.25
Bot settings
| set_my_commands() | Used for setting the list of commands |
| delete_my_commands() | Used for deleting the list of commands |
| get_my_commands() | Used for obtaining the list of commands |
| get_my_default_administrator_rights() | Used for obtaining the default administrator rights for the bot |
| set_my_default_administrator_rights() | Used for setting the default administrator rights for the bot |
| get_chat_menu_button() | Used for obtaining the menu button of a private chat or the default menu button |
| set_chat_menu_button() | Used for setting the menu button of a private chat or the default menu button |
| set_my_description() | Used for setting the description of the bot |
| get_my_description() | Used for obtaining the description of the bot |
| set_my_short_description() | Used for setting the short description of the bot |
| get_my_short_description() | Used for obtaining the short description of the bot |
| set_my_name() | Used for setting the name of the bot |
| get_my_name() | Used for obtaining the name of the bot |
| set_my_profile_photo() | Used for setting the profile photo of the bot |
| remove_my_profile_photo() | Used for removing the profile photo of the bot |
[링크 : https://docs.python-telegram-bot.org/en/stable/telegram.bot.html]
| openclaw agent 구성 관련 (0) | 2026.05.22 |
|---|---|
| mxfp4 (0) | 2026.05.22 |
| openclaw 설치 telegram 신기하네? (0) | 2026.05.22 |
| wan2.2 + comfyui 시도 (0) | 2026.05.21 |
| wan2.2 + comfyui (0) | 2026.05.19 |
설치
| $ curl -fsSL https://openclaw.ai/install.sh | bash |
설정. llama-swap에 의해서 돌고 있는 모델 등록
| $ openclaw onboard --non-interactive \ --auth-choice custom-api-key \ --custom-base-url "http://127.0.0.1:8080/v1" \ --custom-model-id "gemma4-e4b" \ --custom-api-key "llama.cpp" \ --secret-input-mode plaintext \ --custom-compatibility openai \ --accept-risk $ openclaw gateway restart $ openclaw chat --message "안녕, 지금 어떤 모델이 실행 중이야?" |
/ 누르면 아래에 자동완성 처럼 뜨는데 신기하네
그리고 browse providers 처럼 먼가.. 내가 하던거랑은 좀 많이 다른(?) 기능들이 있었나 보다.
텔레그램 봇에 등록되는거라 내걸로 돌려도 메뉴가 계속 뜬다. 젠장!
서비스 종료.
| $ systemctl stop openclaw-gateway --user |
--user를 안하면 없는 서비스라 나온다. 신기하네
| $ systemctl status openclaw-gateway Unit openclaw-gateway.service could not be found. |
| mxfp4 (0) | 2026.05.22 |
|---|---|
| 텔레그램 봇 - 메시지 관련 (0) | 2026.05.22 |
| wan2.2 + comfyui 시도 (0) | 2026.05.21 |
| wan2.2 + comfyui (0) | 2026.05.19 |
| comfyui 실행 (0) | 2026.05.19 |
심심하면(?) VRAM 부족으로 터져서
| $ python3 main.py --listen 0.0.0.0 setup plugin alembic.autogenerate.schemas setup plugin alembic.autogenerate.tables setup plugin alembic.autogenerate.types setup plugin alembic.autogenerate.constraints setup plugin alembic.autogenerate.defaults setup plugin alembic.autogenerate.comments WARNING: You need pytorch with cu130 or higher to use optimized CUDA operations. Found comfy_kitchen backend cuda: {'available': True, 'disabled': True, 'unavailable_reason': None, 'capabilities': ['apply_rope', 'apply_rope1', 'dequantize_nvfp4', 'dequantize_per_tensor_fp8', 'quantize_mxfp8', 'quantize_nvfp4', 'quantize_per_tensor_fp8', 'scaled_mm_nvfp4']} Found comfy_kitchen backend eager: {'available': True, 'disabled': False, 'unavailable_reason': None, 'capabilities': ['apply_rope', 'apply_rope1', 'dequantize_mxfp8', 'dequantize_nvfp4', 'dequantize_per_tensor_fp8', 'quantize_mxfp8', 'quantize_nvfp4', 'quantize_per_tensor_fp8', 'scaled_mm_mxfp8', 'scaled_mm_nvfp4']} Found comfy_kitchen backend triton: {'available': True, 'disabled': True, 'unavailable_reason': None, 'capabilities': ['apply_rope', 'apply_rope1', 'dequantize_nvfp4', 'dequantize_per_tensor_fp8', 'quantize_mxfp8', 'quantize_nvfp4', 'quantize_per_tensor_fp8']} Checkpoint files will always be loaded safely. Total VRAM 11165 MB, total RAM 31755 MB pytorch version: 2.7.1+cu118 Set vram state to: NORMAL_VRAM Device: cuda:0 NVIDIA GeForce GTX 1080 Ti : cudaMallocAsync Using async weight offloading with 2 streams Enabled pinned memory 28579.0 Using pytorch attention Unsupported Pytorch detected. DynamicVRAM support requires Pytorch version 2.8 or later. Falling back to legacy ModelPatcher. VRAM estimates may be unreliable especially on Windows Python version: 3.10.12 (main, Mar 3 2026, 11:56:32) [GCC 11.4.0] ComfyUI version: 0.21.1 comfy-aimdo version: 0.3.0 comfy-kitchen version: 0.2.8 comfyui-frontend-package version: 1.43.18 comfyui-workflow-templates version: 0.9.77 comfyui-embedded-docs version: 0.5.0 comfy-kitchen version: 0.2.8 comfy-aimdo version: 0.3.0 [Prompt Server] web root: /home/minimonk/.local/lib/python3.10/site-packages/comfyui_frontend_package/static Asset seeder disabled Import times for custom nodes: 0.0 seconds: /mnt/Downloads/ComfyUI/custom_nodes/websocket_image_save.py Context impl SQLiteImpl. Will assume non-transactional DDL. Starting server To see the GUI go to: http://0.0.0.0:8188 got prompt Using pytorch attention in VAE Using pytorch attention in VAE VAE load device: cuda:0, offload device: cpu, dtype: torch.float32 Found quantization metadata version 1 Using MixedPrecisionOps for text encoder Requested to load WanTEModel loaded completely; 6419.48 MB loaded, full load: True CLIP/text encoder model load device: cpu, offload device: cpu, current: cpu, dtype: torch.float16 Requested to load WanVAE 0 models unloaded. loaded partially; 0.00 MB usable, 0.00 MB loaded, 484.00 MB offloaded, 45.57 MB buffer reserved, lowvram patches: 0 Found quantization metadata version 1 Detected mixed precision quantization Using mixed precision operations Native ops: , emulated ops: mxfp8, float8_e4m3fn, float8_e5m2, nvfp4 model weight dtype torch.float16, manual cast: torch.float32 model_type FLOW Requested to load WAN21 0 models unloaded. loaded partially; 0.00 MB usable, 0.00 MB loaded, 13636.09 MB offloaded, 885.22 MB buffer reserved, lowvram patches: 0 0%| | 0/10 [00:33<?, ?it/s] !!! Exception during processing !!! Allocation on device Traceback (most recent call last): File "/mnt/Downloads/ComfyUI/execution.py", line 535, in execute output_data, output_ui, has_subgraph, has_pending_tasks = await get_output_data(prompt_id, unique_id, obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, v3_data=v3_data) File "/mnt/Downloads/ComfyUI/execution.py", line 335, in get_output_data return_values = await _async_map_node_over_list(prompt_id, unique_id, obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, v3_data=v3_data) File "/mnt/Downloads/ComfyUI/execution.py", line 309, in _async_map_node_over_list await process_inputs(input_dict, i) File "/mnt/Downloads/ComfyUI/execution.py", line 297, in process_inputs result = f(**inputs) File "/mnt/Downloads/ComfyUI/nodes.py", line 1612, in sample return common_ksampler(model, noise_seed, steps, cfg, sampler_name, scheduler, positive, negative, latent_image, denoise=denoise, disable_noise=disable_noise, start_step=start_at_step, last_step=end_at_step, force_full_denoise=force_full_denoise) File "/mnt/Downloads/ComfyUI/nodes.py", line 1542, in common_ksampler samples = comfy.sample.sample(model, noise, steps, cfg, sampler_name, scheduler, positive, negative, latent_image, File "/mnt/Downloads/ComfyUI/comfy/sample.py", line 74, in sample samples = sampler.sample(noise, positive, negative, cfg=cfg, latent_image=latent_image, start_step=start_step, last_step=last_step, force_full_denoise=force_full_denoise, denoise_mask=noise_mask, sigmas=sigmas, callback=callback, disable_pbar=disable_pbar, seed=seed) File "/mnt/Downloads/ComfyUI/comfy/samplers.py", line 1180, in sample return sample(self.model, noise, positive, negative, cfg, self.device, sampler, sigmas, self.model_options, latent_image=latent_image, denoise_mask=denoise_mask, callback=callback, disable_pbar=disable_pbar, seed=seed) File "/mnt/Downloads/ComfyUI/comfy/samplers.py", line 1070, in sample return cfg_guider.sample(noise, latent_image, sampler, sigmas, denoise_mask, callback, disable_pbar, seed) File "/mnt/Downloads/ComfyUI/comfy/samplers.py", line 1052, in sample output = executor.execute(noise, latent_image, sampler, sigmas, denoise_mask, callback, disable_pbar, seed, latent_shapes=latent_shapes) File "/mnt/Downloads/ComfyUI/comfy/patcher_extension.py", line 112, in execute return self.original(*args, **kwargs) File "/mnt/Downloads/ComfyUI/comfy/samplers.py", line 995, in outer_sample output = self.inner_sample(noise, latent_image, device, sampler, sigmas, denoise_mask, callback, disable_pbar, seed, latent_shapes=latent_shapes) File "/mnt/Downloads/ComfyUI/comfy/samplers.py", line 981, in inner_sample samples = executor.execute(self, sigmas, extra_args, callback, noise, latent_image, denoise_mask, disable_pbar) File "/mnt/Downloads/ComfyUI/comfy/patcher_extension.py", line 112, in execute return self.original(*args, **kwargs) File "/mnt/Downloads/ComfyUI/comfy/samplers.py", line 751, in sample samples = self.sampler_function(model_k, noise, sigmas, extra_args=extra_args, callback=k_callback, disable=disable_pbar, **self.extra_options) File "/home/minimonk/.local/lib/python3.10/site-packages/torch/utils/_contextlib.py", line 116, in decorate_context return func(*args, **kwargs) File "/mnt/Downloads/ComfyUI/comfy/k_diffusion/sampling.py", line 205, in sample_euler denoised = model(x, sigma_hat * s_in, **extra_args) File "/mnt/Downloads/ComfyUI/comfy/samplers.py", line 400, in __call__ out = self.inner_model(x, sigma, model_options=model_options, seed=seed) File "/mnt/Downloads/ComfyUI/comfy/samplers.py", line 954, in __call__ return self.outer_predict_noise(*args, **kwargs) File "/mnt/Downloads/ComfyUI/comfy/samplers.py", line 961, in outer_predict_noise ).execute(x, timestep, model_options, seed) File "/mnt/Downloads/ComfyUI/comfy/patcher_extension.py", line 112, in execute return self.original(*args, **kwargs) File "/mnt/Downloads/ComfyUI/comfy/samplers.py", line 964, in predict_noise return sampling_function(self.inner_model, x, timestep, self.conds.get("negative", None), self.conds.get("positive", None), self.cfg, model_options=model_options, seed=seed) File "/mnt/Downloads/ComfyUI/comfy/samplers.py", line 380, in sampling_function out = calc_cond_batch(model, conds, x, timestep, model_options) File "/mnt/Downloads/ComfyUI/comfy/samplers.py", line 205, in calc_cond_batch return _calc_cond_batch_outer(model, conds, x_in, timestep, model_options) File "/mnt/Downloads/ComfyUI/comfy/samplers.py", line 213, in _calc_cond_batch_outer return executor.execute(model, conds, x_in, timestep, model_options) File "/mnt/Downloads/ComfyUI/comfy/patcher_extension.py", line 112, in execute return self.original(*args, **kwargs) File "/mnt/Downloads/ComfyUI/comfy/samplers.py", line 325, in _calc_cond_batch output = model.apply_model(input_x, timestep_, **c).chunk(batch_chunks) File "/mnt/Downloads/ComfyUI/comfy/model_base.py", line 182, in apply_model return comfy.patcher_extension.WrapperExecutor.new_class_executor( File "/mnt/Downloads/ComfyUI/comfy/patcher_extension.py", line 112, in execute return self.original(*args, **kwargs) File "/mnt/Downloads/ComfyUI/comfy/model_base.py", line 226, in _apply_model model_output = self.diffusion_model(xc, t, context=context, control=control, transformer_options=transformer_options, **extra_conds) File "/home/minimonk/.local/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1751, in _wrapped_call_impl return self._call_impl(*args, **kwargs) File "/home/minimonk/.local/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1762, in _call_impl return forward_call(*args, **kwargs) File "/mnt/Downloads/ComfyUI/comfy/ldm/wan/model.py", line 644, in forward return comfy.patcher_extension.WrapperExecutor.new_class_executor( File "/mnt/Downloads/ComfyUI/comfy/patcher_extension.py", line 112, in execute return self.original(*args, **kwargs) File "/mnt/Downloads/ComfyUI/comfy/ldm/wan/model.py", line 664, in _forward return self.forward_orig(x, timestep, context, clip_fea=clip_fea, freqs=freqs, transformer_options=transformer_options, **kwargs)[:, :, :t, :h, :w] File "/mnt/Downloads/ComfyUI/comfy/ldm/wan/model.py", line 597, in forward_orig x = block(x, e=e0, freqs=freqs, context=context, context_img_len=context_img_len, transformer_options=transformer_options) File "/home/minimonk/.local/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1751, in _wrapped_call_impl return self._call_impl(*args, **kwargs) File "/home/minimonk/.local/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1762, in _call_impl return forward_call(*args, **kwargs) File "/mnt/Downloads/ComfyUI/comfy/ldm/wan/model.py", line 258, in forward y = self.ffn(torch.addcmul(repeat_e(e[3], x), self.norm2(x), 1 + repeat_e(e[4], x))) File "/home/minimonk/.local/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1751, in _wrapped_call_impl return self._call_impl(*args, **kwargs) File "/home/minimonk/.local/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1762, in _call_impl return forward_call(*args, **kwargs) File "/home/minimonk/.local/lib/python3.10/site-packages/torch/nn/modules/container.py", line 240, in forward input = module(input) File "/home/minimonk/.local/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1751, in _wrapped_call_impl return self._call_impl(*args, **kwargs) File "/home/minimonk/.local/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1762, in _call_impl return forward_call(*args, **kwargs) File "/home/minimonk/.local/lib/python3.10/site-packages/torch/nn/modules/activation.py", line 734, in forward return F.gelu(input, approximate=self.approximate) torch.OutOfMemoryError: Allocation on device Memory summary: |===========================================================================| | PyTorch CUDA memory summary, device ID 0 | |---------------------------------------------------------------------------| | CUDA OOMs: 0 | cudaMalloc retries: 0 | |===========================================================================| | Metric | Cur Usage | Peak Usage | Tot Alloc | Tot Freed | |---------------------------------------------------------------------------| | Allocated memory | 5675 MiB | 7030 MiB | 0 B | 0 B | | from large pool | 0 MiB | 0 MiB | 0 B | 0 B | | from small pool | 0 MiB | 0 MiB | 0 B | 0 B | |---------------------------------------------------------------------------| | Active memory | 5675 MiB | 7030 MiB | 0 B | 0 B | | from large pool | 0 MiB | 0 MiB | 0 B | 0 B | | from small pool | 0 MiB | 0 MiB | 0 B | 0 B | |---------------------------------------------------------------------------| | Requested memory | 0 B | 0 B | 0 B | 0 B | | from large pool | 0 B | 0 B | 0 B | 0 B | | from small pool | 0 B | 0 B | 0 B | 0 B | |---------------------------------------------------------------------------| | GPU reserved memory | 10784 MiB | 10784 MiB | 0 B | 0 B | | from large pool | 0 MiB | 0 MiB | 0 B | 0 B | | from small pool | 0 MiB | 0 MiB | 0 B | 0 B | |---------------------------------------------------------------------------| | Non-releasable memory | 0 B | 0 B | 0 B | 0 B | | from large pool | 0 B | 0 B | 0 B | 0 B | | from small pool | 0 B | 0 B | 0 B | 0 B | |---------------------------------------------------------------------------| | Allocations | 0 | 0 | 0 | 0 | | from large pool | 0 | 0 | 0 | 0 | | from small pool | 0 | 0 | 0 | 0 | |---------------------------------------------------------------------------| | Active allocs | 0 | 0 | 0 | 0 | | from large pool | 0 | 0 | 0 | 0 | | from small pool | 0 | 0 | 0 | 0 | |---------------------------------------------------------------------------| | GPU reserved segments | 0 | 0 | 0 | 0 | | from large pool | 0 | 0 | 0 | 0 | | from small pool | 0 | 0 | 0 | 0 | |---------------------------------------------------------------------------| | Non-releasable allocs | 0 | 0 | 0 | 0 | | from large pool | 0 | 0 | 0 | 0 | | from small pool | 0 | 0 | 0 | 0 | |---------------------------------------------------------------------------| | Oversize allocations | 0 | 0 | 0 | 0 | |---------------------------------------------------------------------------| | Oversize GPU segments | 0 | 0 | 0 | 0 | |===========================================================================| Got an OOM, unloading all loaded models. Prompt executed in 154.15 seconds |
둘 중에 하나 주면 된다는데 해도 터지고
| python3 main.py --listen 0.0.0.0 --lowvram python3 main.py --listen 0.0.0.0 --novram |
멀 하다가 꼬였는지 패키지 문제가 생겨서 다시 밀고 cuda 11.8에 맞춰서 재설치
| pip3 uninstall -y torch torchvision torchaudio xformers pip3 install torch==2.6.0 torchvision torchaudio --index-url https://download.pytorch.org/whl/cu118 pip3 install xformers==0.0.29.post2 |
그래도 하려니 터져서 768x768 이었던게 왜 800x800이 되었는진 모르겠고 길이가 81로 되어있어서
일단은 512x512에 29 로 바꾸고 재시도
먼가 되는거 같긴한데.. offloaded가 엄청 크네.
| got prompt Requested to load WanVAE loaded completely; 4417.69 MB usable, 484.06 MB loaded, full load: True [MultiGPU Runtime] Using runtime device cuda:0 (comfy.sample.sample:ModelPatcher) Requested to load WAN21 loaded partially; 7397.49 MB usable, 7211.06 MB loaded, 6425.03 MB offloaded, 175.06 MB buffer reserved, lowvram patches: 0 0%| | 0/10 [00:00<?, ?it/s] 20%|████████▌ | 2/10 [09:15<37:03, 277.94s/it] |
1시간은 족히 넘은거 같은데 (37 띄워놓고 거짓말 쟁이!)
이제야 두번째 KSampler로 넘어갔다!!! 이예!!!
메모리 답이 안나올거 같아서 low_noise가 아닌 high_noise 에 해상도까지 낮추고 했는데
1시간 33분 16 동안 해서 29frame / 16fps 해서 대충 1.8초 짜리 똥을 생성해냄
| loaded completely; 883.22 MB usable, 484.06 MB loaded, full load: True Prompt executed in 01:33:16 |
이게 머야 ㅋㅋㅋ
+
왜 이번에는 webm이 아니라 webp일까?
그 와중에 49frame / 16fps 약 3초 만드는데 2시간 30분..
2시간 30분 짜리 똥이야!
| got prompt Found quantization metadata version 1 Detected mixed precision quantization Using mixed precision operations Native ops: , emulated ops: float8_e4m3fn, mxfp8, nvfp4, float8_e5m2 model weight dtype torch.float16, manual cast: torch.float32 model_type FLOW [MultiGPU Runtime] Using runtime device cuda:0 (comfy.sample.sample:ModelPatcher) Requested to load WAN21 loaded partially; 5509.86 MB usable, 5037.30 MB loaded, 8598.79 MB offloaded, 472.56 MB buffer reserved, lowvram patches: 0 100%|████████████████████████████████████████| 10/10 [1:14:32<00:00, 447.24s/it] [MultiGPU Runtime] Using runtime device cuda:0 (comfy.sample.sample:ModelPatcher) Requested to load WAN21 loaded partially; 5495.86 MB usable, 5019.61 MB loaded, 8616.47 MB offloaded, 472.56 MB buffer reserved, lowvram patches: 0 0%| | 0/10 [00:00<?, ?it/s]100%|████████████████████████████████████████| 10/10 [1:14:33<00:00, 447.32s/it] Requested to load WanVAE Unloaded partially: 2187.40 MB freed, 2832.21 MB remains loaded, 472.56 MB buffer reserved, lowvram patches: 0 loaded completely; 545.46 MB usable, 484.06 MB loaded, full load: True Prompt executed in 02:30:07 |
| 텔레그램 봇 - 메시지 관련 (0) | 2026.05.22 |
|---|---|
| openclaw 설치 telegram 신기하네? (0) | 2026.05.22 |
| wan2.2 + comfyui (0) | 2026.05.19 |
| comfyui 실행 (0) | 2026.05.19 |
| llama.cpp kv quantization (0) | 2026.05.19 |
일단 여기꺼 보고 시작함
[링크 : https://marcus-story.tistory.com/239]
[링크 : https://huggingface.co/Comfy-Org/Wan_2.2_ComfyUI_Repackaged/tree/main/split_files/text_encoders]
[링크 : https://huggingface.co/Comfy-Org/Wan_2.2_ComfyUI_Repackaged/tree/main/split_files/diffusion_models]
[링크 : https://huggingface.co/Comfy-Org/Wan_2.2_ComfyUI_Repackaged/tree/main/split_files/vae]
[링크 : https://huggingface.co/Comfy-Org/Wan_2.2_ComfyUI_Repackaged/tree/main]
[링크 : https://comfyanonymous.github.io/ComfyUI_examples/wan22/]
[링크 : https://github.com/Wan-Video/Wan2.2]
[링크 : https://huggingface.co/Wan-AI/Wan2.2-I2V-A14B/tree/main]
[링크 : https://huggingface.co/Wan-AI/Wan2.2-TI2V-5B/tree/main]
[링크 : https://huggingface.co/Wan-AI/Wan2.2-T2V-A14B/tree/main]
json 파일 저장하고 그냥 드래그 해서 던지면 불러온다. 붙여 넣거나 하는건 안되나?
하나씩 모델 다운로드 하고 해봐야지 머 ㅋㅋ
| openclaw 설치 telegram 신기하네? (0) | 2026.05.22 |
|---|---|
| wan2.2 + comfyui 시도 (0) | 2026.05.21 |
| comfyui 실행 (0) | 2026.05.19 |
| llama.cpp kv quantization (0) | 2026.05.19 |
| openai api (0) | 2026.05.18 |
| /mnt/Downloads/ComfyUI$ python3 main.py --listen 0.0.0.0 setup plugin alembic.autogenerate.schemas setup plugin alembic.autogenerate.tables setup plugin alembic.autogenerate.types setup plugin alembic.autogenerate.constraints setup plugin alembic.autogenerate.defaults setup plugin alembic.autogenerate.comments WARNING: You need pytorch with cu130 or higher to use optimized CUDA operations. Found comfy_kitchen backend triton: {'available': True, 'disabled': True, 'unavailable_reason': None, 'capabilities': ['apply_rope', 'apply_rope1', 'dequantize_nvfp4', 'dequantize_per_tensor_fp8', 'quantize_mxfp8', 'quantize_nvfp4', 'quantize_per_tensor_fp8']} Found comfy_kitchen backend cuda: {'available': True, 'disabled': True, 'unavailable_reason': None, 'capabilities': ['apply_rope', 'apply_rope1', 'dequantize_nvfp4', 'dequantize_per_tensor_fp8', 'quantize_mxfp8', 'quantize_nvfp4', 'quantize_per_tensor_fp8', 'scaled_mm_nvfp4']} Found comfy_kitchen backend eager: {'available': True, 'disabled': False, 'unavailable_reason': None, 'capabilities': ['apply_rope', 'apply_rope1', 'dequantize_mxfp8', 'dequantize_nvfp4', 'dequantize_per_tensor_fp8', 'quantize_mxfp8', 'quantize_nvfp4', 'quantize_per_tensor_fp8', 'scaled_mm_mxfp8', 'scaled_mm_nvfp4']} Checkpoint files will always be loaded safely. Total VRAM 11165 MB, total RAM 15651 MB pytorch version: 2.7.1+cu118 Set vram state to: NORMAL_VRAM Device: cuda:0 NVIDIA GeForce GTX 1080 Ti : cudaMallocAsync Using async weight offloading with 2 streams Enabled pinned memory 14085.0 Using pytorch attention Unsupported Pytorch detected. DynamicVRAM support requires Pytorch version 2.8 or later. Falling back to legacy ModelPatcher. VRAM estimates may be unreliable especially on WindowsPython version: 3.10.12 (main, Mar 3 2026, 11:56:32) [GCC 11.4.0] ComfyUI version: 0.21.1 comfy-aimdo version: 0.3.0 comfy-kitchen version: 0.2.8 comfyui-frontend-package version: 1.43.18 comfyui-workflow-templates version: 0.9.77 comfyui-embedded-docs version: 0.5.0 comfy-kitchen version: 0.2.8 comfy-aimdo version: 0.3.0 [Prompt Server] web root: /home/minimonk/.local/lib/python3.10/site-packages/comfyui_frontend_package/static Asset seeder disabled Import times for custom nodes: 0.0 seconds: /mnt/Downloads/ComfyUI/custom_nodes/websocket_image_save.py Context impl SQLiteImpl. Will assume non-transactional DDL. Starting server To see the GUI go to: http://0.0.0.0:8188 ^C Stopped server |
그냥 켜면 기본으로 있는데
몰라서(!) ComfyUI/models/diffusion_models 에 *.safetensors 파일들을 넣어놓고는 체크포인트 로드에 왜 안뜨나 했는데
ComfyUI/models/checkpoints 에 넣어주고 리프레시 하면 뜬다.
[링크 : https://comfyui-wiki.com/ko/comfyui-nodes/loaders/checkpoint-loader-simple]
선이 먼가 드럽게 꼬여서
정리 하는데. 도대체 저 하나하나 노드들을 멀로 추가해야하나 모르겠다.
걍 CLIP 부분 드래그 해서 오니 추천으로 "CLIP 테스트 인코딩 (프롬프트)" 가 뜨니
우클릭해서 먼지 찾을 필요가 없을 듯?
굳이.. 하겠다면
노드 추가 - 조건화 - CLIP 텍스트 인코딩 (프롬프트) 로 하면 될 것같은데
CLIP이 먼지 찾아봐야겠다.
+
[링크 : https://youngri.tistory.com/m/40/] api 서버?
[링크 : https://m.blog.naver.com/minwoo932/224142522248]
[링크 : https://comfyui-wiki.com/ko/tutorial/basic/creating-your-first-image-by-the-first-time]
| wan2.2 + comfyui 시도 (0) | 2026.05.21 |
|---|---|
| wan2.2 + comfyui (0) | 2026.05.19 |
| llama.cpp kv quantization (0) | 2026.05.19 |
| openai api (0) | 2026.05.18 |
| RAG 시도 - postgresql(14) + pgvector (1) | 2026.05.15 |
128k context length에 2기가 VRAM을 냠냠
| $ ./llama-b8925/llama-cli -m model/gemma4-e4b/gemma-4-E4B-it-Q4_K_M.gguf --verbose llama_kv_cache: size = 2048.00 MiB (131072 cells, 4 layers, 1/1 seqs), K (f16): 1024.00 MiB, V (f16): 1024.00 MiB llama_kv_cache: attn_rot_k = 0, n_embd_head_k_all = 512 llama_kv_cache: attn_rot_v = 0, n_embd_head_k_all = 512 llama_kv_cache_iswa: creating SWA KV cache, size = 1024 cells |
k 만 q4로 하니 288MB! 256MB 보단 약간 큰데 아무튼 대충~ 1/4 로 줄었다.
| $ ./llama-b8925/llama-cli -m model/gemma4-e4b/gemma-4-E4B-it-Q4_K_M.gguf --verbose -ctk q4_0 --ctx-size 131072 llama_kv_cache: size = 1312.00 MiB (131072 cells, 4 layers, 1/1 seqs), K (q4_0): 288.00 MiB, V (f16): 1024.00 MiB llama_kv_cache: attn_rot_k = 1, n_embd_head_k_all = 512 llama_kv_cache: attn_rot_v = 0, n_embd_head_k_all = 512 llama_kv_cache_iswa: creating SWA KV cache, size = 1024 cells |
kv를 q4로 하니 대충 512MB 근처.
| $ ./llama-b8925/llama-cli -m model/gemma4-e4b/gemma-4-E4B-it-Q4_K_M.gguf --verbose -ctk q4_0 -ctv q4_0 --ctx-size 131072 llama_kv_cache: size = 576.00 MiB (131072 cells, 4 layers, 1/1 seqs), K (q4_0): 288.00 MiB, V (q4_0): 288.00 MiB llama_kv_cache: attn_rot_k = 1, n_embd_head_k_all = 512 llama_kv_cache: attn_rot_v = 1, n_embd_head_k_all = 512 llama_kv_cache_iswa: creating SWA KV cache, size = 1024 cells |
+
2026.05.29
verbose로는 토큰별로 로그가 나와서 성능 저하가 있긴한데
| verbose | none | ctk | ctk ctv | ctv |
| 단문 | 50.13 t/s | 58.27 t/s | 43.67 t/s | 51.84 t/s |
| 중문 | 50.76 t/s | 56.41 t/s | 43.66 t/s | 50.37 t/s |
| 장문 | 49.70 t/s | 54.66 t/s | 42.75 t/s | 45.81 t/s |
생각외로 양자화 한거랑 안한거랑 차이가 별로 없다.
1080 이라 q4를 지원하지 않아서 그런걸지도?
| - | none | ctk | ctk ctv | ctv |
| 단문 | 57.48 t/s | 57.03 t/s | 51.67 t/s | 52.35 t/s |
| 중문 | 59.56 t/s | 56.90 t/s | 50.20 t/s | 49.95 t/s |
| 장문 | 53.83 t/s | 54.82 t/s | 46.28 t/s | 44.93 t/s |
안녕?
너에 대해서 소개해줘
파이썬으로 셀레니움을 통해 웹을 서칭하고 텍스트만 추출하고 makrdown 으로 변환후 md 파일과 pdf로 저장하는 기능을 구현해줘
/mnt/Downloads/llama-b9305/llama-server --model /mnt/Downloads/model/gemma4-e4b/gemma-4-E4B-it-Q4_K_M.gguf -mm ./model/gemma4-e4b/mmproj-F16.gguf -sm none --ctx-size 131072 --verbose --host 0.0.0.0
gemma-4-E4B-it-Q4_K_M.gguf
Reading
Generation
63 tokens
1.3s
50.13 t/s
gemma-4-E4B-it-Q4_K_M.gguf
Reading
Generation
875 tokens
17s
50.76 t/s
gemma-4-E4B-it-Q4_K_M.gguf
Reading
Generation
3,009 tokens
1min
49.70 t/s
/mnt/Downloads/llama-b9305/llama-server --model /mnt/Downloads/model/gemma4-e4b/gemma-4-E4B-it-Q4_K_M.gguf -mm ./model/gemma4-e4b/mmproj-F16.gguf -sm none -ctk q4_0 --ctx-size 131072 --verbose --host 0.0.0.0
gemma-4-E4B-it-Q4_K_M.gguf
Reading
Generation
41 tokens
0.7s
58.27 t/s
gemma-4-E4B-it-Q4_K_M.gguf
Reading
Generation
1,025 tokens
18s
56.41 t/s
gemma-4-E4B-it-Q4_K_M.gguf
Reading
Generation
3,098 tokens
56s
54.66 t/s
/mnt/Downloads/llama-b9305/llama-server --model /mnt/Downloads/model/gemma4-e4b/gemma-4-E4B-it-Q4_K_M.gguf -mm ./model/gemma4-e4b/mmproj-F16.gguf -sm none -ctk q4_0 -ctv q4_0 --ctx-size 131072 --verbose --host 0.0.0.0
gemma-4-E4B-it-Q4_K_M.gguf
Reading
Generation
63 tokens
1.4s
43.67 t/s
gemma-4-E4B-it-Q4_K_M.gguf
Reading
Generation
1,059 tokens
24s
43.66 t/s
gemma-4-E4B-it-Q4_K_M.gguf
Reading
Generation
2,558 tokens
59s
42.75 t/s
/mnt/Downloads/llama-b9305/llama-server --model /mnt/Downloads/model/gemma4-e4b/gemma-4-E4B-it-Q4_K_M.gguf -mm ./model/gemma4-e4b/mmproj-F16.gguf -sm none -ctv q4_0 --ctx-size 131072 --verbose --host 0.0.0.0
gemma-4-E4B-it-Q4_K_M.gguf
Reading
Generation
328 tokens
6.3s
51.84 t/s
gemma-4-E4B-it-Q4_K_M.gguf
Reading
Generation
902 tokens
17s
50.37 t/s
gemma-4-E4B-it-Q4_K_M.gguf
Reading
Generation
3,105 tokens
1min 7s
45.81 t/s
---------------------
$ /mnt/Downloads/llama-b9305/llama-server --model /mnt/Downloads/model/gemma4-e4b/gemma-4-E4B-it-Q4_K_M.gguf -mm ./model/gemma4-e4b/mmproj-F16.gguf -sm none -ctv q4_0 --ctx-size 131072 --host 0.0.0.0
gemma-4-E4B-it-Q4_K_M.gguf
Reading
Generation
609 tokens
11s
52.35 t/s
gemma-4-E4B-it-Q4_K_M.gguf
Reading
Generation
850 tokens
17s
49.95 t/s
gemma-4-E4B-it-Q4_K_M.gguf
Reading
Generation
3,472 tokens
1min 17s
44.93 t/s
$ /mnt/Downloads/llama-b9305/llama-server --model /mnt/Downloads/model/gemma4-e4b/gemma-4-E4B-it-Q4_K_M.gguf -mm ./model/gemma4-e4b/mmproj-F16.gguf -sm none -ctk q4_0 --ctx-size 131072 --host 0.0.0.0
gemma-4-E4B-it-Q4_K_M.gguf
Reading
Generation
32 tokens
0.6s
57.03 t/s
gemma-4-E4B-it-Q4_K_M.gguf
Reading
Generation
640 tokens
11s
56.90 t/s
gemma-4-E4B-it-Q4_K_M.gguf
Reading
Generation
3,379 tokens
1min 1s
54.82 t/s
$ /mnt/Downloads/llama-b9305/llama-server --model /mnt/Downloads/model/gemma4-e4b/gemma-4-E4B-it-Q4_K_M.gguf -mm ./model/gemma4-e4b/mmproj-F16.gguf -sm none -ctk q4_0 -ctv q4_0 --ctx-size 131072 --host 0.0.0.0
gemma-4-E4B-it-Q4_K_M.gguf
Reading
Generation
37 tokens
0.7s
51.67 t/s
gemma-4-E4B-it-Q4_K_M.gguf
Reading
Generation
817 tokens
16s
50.20 t/s
gemma-4-E4B-it-Q4_K_M.gguf
Reading
Generation
3,357 tokens
1min 12s
46.28 t/s
$ /mnt/Downloads/llama-b9305/llama-server --model /mnt/Downloads/model/gemma4-e4b/gemma-4-E4B-it-Q4_K_M.gguf -mm ./model/gemma4-e4b/mmproj-F16.gguf -sm none --ctx-size 131072 --host 0.0.0.0
gemma-4-E4B-it-Q4_K_M.gguf
Reading
Generation
45 tokens
0.8s
57.48 t/s
gemma-4-E4B-it-Q4_K_M.gguf
Reading
Generation
737 tokens
12s
59.56 t/s
gemma-4-E4B-it-Q4_K_M.gguf
Reading
Generation
3,291 tokens
1min 1s
53.83 t/s
| wan2.2 + comfyui (0) | 2026.05.19 |
|---|---|
| comfyui 실행 (0) | 2026.05.19 |
| openai api (0) | 2026.05.18 |
| RAG 시도 - postgresql(14) + pgvector (1) | 2026.05.15 |
| gpt님 만세! - pip torch 버전 낮추기 (0) | 2026.05.15 |
[링크 : https://velog.io/@euisuk-chung/%EC%A0%95%EB%A6%AC-OpenAI-API-Document]
[링크 : https://developers.openai.com/api/reference/resources/responses]
[링크 : https://developers.openai.com/api/docs/guides/text]
[링크 : https://model-spec.openai.com/2025-02-12.html#chain_of_command]
claude에 물어보니 openai 의 api가 랩핑한것일뿐
결국에는 신경망에 특수 토큰으로 인식될 만한 녀석으로 감싸고 그걸 임베딩 해서 벡터로 넣어줄 뿐
무언가(?) 대단한 방법으로 들어가는건 아니라고 한다. imagenary monologue 줄여서 im prefix
| ChatML (Qwen, Hermes) | <|im_start|>role | <|im_end|> | 범용 role 삽입 |
| LLaMA-3 | <|start_header_id|>role<|end_header_id|> | <|eot_id|> | 헤더 분리 |
| LLaMA-2 | [INST] | [/INST] | role 구분 없음 |
| Mistral | [INST] | [/INST] | role 구분 없음 |
| DeepSeek-R1 | <|User|> / <|Assistant|> | <|end▁of▁sentence|> | role별 고정 토큰 |
[링크 : https://claude.ai/share/0bfc9c76-e5a8-46d2-8346-4f84170625db]
| #define CHATML_TEMPLATE_SRC \ "{%- for message in messages -%}\n" \ " {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>\n' -}}\n" \ "{%- endfor -%}\n" \ "{%- if add_generation_prompt -%}\n" \ " {{- '<|im_start|>assistant\n' -}}\n" \ "{%- endif -%}" const std::string TOOL_CALL_START = "<|tool_call_start|>"; const std::string TOOL_CALL_END = "<|tool_call_end|>"; const std::string THINK_START = "<think>"; const std::string THINK_END = "</think>"; const std::string GEN_PROMPT = "<|im_start|>assistant\n"; |
[링크 : https://github.com/ggml-org/llama.cpp/blob/master/common/chat.cpp#L567]
음.. gemma4-e4b 에서는 im_start가 특수 토큰으로 배우진 않은듯?
| /mnt/Downloads/model/gemma4-e4b$ ../../llama-b8925/llama-tokenize -m gemma-4-E4B-it-Q4_K_M.gguf -p "<|im_start|>system" --log-disable 2 -> '<bos>' 236820 -> '<' 236909 -> '|' 548 -> 'im' 236779 -> '_' 3041 -> 'start' 111038 -> '|>' 9731 -> 'system' /mnt/Downloads/model/gemma4-e4b$ ../../llama-b8925/llama-tokenize -m gemma-4-E4B-it-Q4_K_M.gguf -p "<|im_start|>system" --log-disable --no-parse-special 2 -> '<bos>' 236820 -> '<' 236909 -> '|' 548 -> 'im' 236779 -> '_' 3041 -> 'start' 111038 -> '|>' 9731 -> 'system' /mnt/Downloads/model/gemma4-e4b$ ../../llama-b8925/llama-tokenize -m gemma-4-E4B-it-Q4_K_M.gguf -p "<|im_start|>system" --log-disable --no-parse-special --ids [2, 236820, 236909, 548, 236779, 3041, 111038, 9731] |
+
모델 읽어들이다가 발견 한 특수 토큰..?
일단은 control token 이라고 나온다.
| $ ./llama-b8925/llama-cli -m model/gemma4-e4b/gemma-4-E4B-it-Q4_K_M.gguf --verbose load: control token: 258884 '<|video|>' is not marked as EOG load: control token: 255999 '<|image>' is not marked as EOG load: control token: 258882 '<image|>' is not marked as EOG load: control token: 258883 '<audio|>' is not marked as EOG load: control token: 98 '<|think|>' is not marked as EOG load: control token: 105 '<|turn>' is not marked as EOG load: control token: 258880 '<|image|>' is not marked as EOG load: control token: 2 '<bos>' is not marked as EOG load: control-looking token: 212 '</s>' was not control-type; this is probably a bug in the model. its type will be overridden load: control token: 0 '<pad>' is not marked as EOG load: control-looking token: 50 '<|tool_response>' was not control-type; this is probably a bug in the model. its type will be overridden load: control token: 46 '<|tool>' is not marked as EOG load: control token: 47 '<tool|>' is not marked as EOG load: control token: 256000 '<|audio>' is not marked as EOG load: control token: 3 '<unk>' is not marked as EOG load: control token: 258881 '<|audio|>' is not marked as EOG load: control-looking token: 1 '<eos>' was not control-type; this is probably a bug in the model. its type will be overridden load: control token: 4 '<mask>' is not marked as EOG |load: printing all EOG tokens: load: - 1 ('<eos>') load: - 50 ('<|tool_response>') load: - 106 ('<turn|>') load: - 212 ('</s>') load: special_eog_ids contains '<|tool_response>', removing '</s>' token from EOG list print_info: BOS token = 2 '<bos>' print_info: EOS token = 106 '<turn|>' print_info: UNK token = 3 '<unk>' print_info: PAD token = 0 '<pad>' print_info: MASK token = 4 '<mask>' print_info: LF token = 107 ' ' print_info: EOG token = 1 '<eos>' print_info: EOG token = 50 '<|tool_response>' print_info: EOG token = 106 '<turn|>' print_info: max token length = 93 init: chat template, example_format: '<|turn>system <|think|> You are a helpful assistant<turn|> <|turn>user Hello<turn|> <|turn>model Hi there<turn|> <|turn>user How are you?<turn|> <|turn>model ' srv init: init: chat template, thinking = 1 |
| comfyui 실행 (0) | 2026.05.19 |
|---|---|
| llama.cpp kv quantization (0) | 2026.05.19 |
| RAG 시도 - postgresql(14) + pgvector (1) | 2026.05.15 |
| gpt님 만세! - pip torch 버전 낮추기 (0) | 2026.05.15 |
| wan2.2 (0) | 2026.05.13 |
|
