Instructions to use unsloth/Muse-Glimmer-30B-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use unsloth/Muse-Glimmer-30B-GGUF with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("image-text-to-text", model="unsloth/Muse-Glimmer-30B-GGUF")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("unsloth/Muse-Glimmer-30B-GGUF", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use unsloth/Muse-Glimmer-30B-GGUF with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf unsloth/Muse-Glimmer-30B-GGUF:UD-Q4_K_XL
# Run inference directly in the terminal:
llama cli -hf unsloth/Muse-Glimmer-30B-GGUF:UD-Q4_K_XL

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf unsloth/Muse-Glimmer-30B-GGUF:UD-Q4_K_XL
# Run inference directly in the terminal:
llama cli -hf unsloth/Muse-Glimmer-30B-GGUF:UD-Q4_K_XL

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf unsloth/Muse-Glimmer-30B-GGUF:UD-Q4_K_XL
# Run inference directly in the terminal:
./llama-cli -hf unsloth/Muse-Glimmer-30B-GGUF:UD-Q4_K_XL

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf unsloth/Muse-Glimmer-30B-GGUF:UD-Q4_K_XL
# Run inference directly in the terminal:
./build/bin/llama-cli -hf unsloth/Muse-Glimmer-30B-GGUF:UD-Q4_K_XL

Use Docker

docker model run hf.co/unsloth/Muse-Glimmer-30B-GGUF:UD-Q4_K_XL

LM Studio
Jan

vLLM

How to use unsloth/Muse-Glimmer-30B-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "unsloth/Muse-Glimmer-30B-GGUF"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "unsloth/Muse-Glimmer-30B-GGUF",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/unsloth/Muse-Glimmer-30B-GGUF:UD-Q4_K_XL

SGLang

How to use unsloth/Muse-Glimmer-30B-GGUF with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "unsloth/Muse-Glimmer-30B-GGUF" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "unsloth/Muse-Glimmer-30B-GGUF",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "unsloth/Muse-Glimmer-30B-GGUF" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "unsloth/Muse-Glimmer-30B-GGUF",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Ollama
How to use unsloth/Muse-Glimmer-30B-GGUF with Ollama:
```
ollama run hf.co/unsloth/Muse-Glimmer-30B-GGUF:UD-Q4_K_XL
```

Unsloth Studio

How to use unsloth/Muse-Glimmer-30B-GGUF with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for unsloth/Muse-Glimmer-30B-GGUF to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for unsloth/Muse-Glimmer-30B-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for unsloth/Muse-Glimmer-30B-GGUF to start chatting

How to use unsloth/Muse-Glimmer-30B-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf unsloth/Muse-Glimmer-30B-GGUF:UD-Q4_K_XL

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "unsloth/Muse-Glimmer-30B-GGUF:UD-Q4_K_XL"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use unsloth/Muse-Glimmer-30B-GGUF with Docker Model Runner:
```
docker model run hf.co/unsloth/Muse-Glimmer-30B-GGUF:UD-Q4_K_XL
```

Lemonade

How to use unsloth/Muse-Glimmer-30B-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull unsloth/Muse-Glimmer-30B-GGUF:UD-Q4_K_XL

Run and chat with the model

lemonade run user.Muse-Glimmer-30B-GGUF-UD-Q4_K_XL

List all available models

lemonade list

Hermes Agent

How to use unsloth/Muse-Glimmer-30B-GGUF with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf unsloth/Muse-Glimmer-30B-GGUF:UD-Q4_K_XL

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default unsloth/Muse-Glimmer-30B-GGUF:UD-Q4_K_XL

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use unsloth/Muse-Glimmer-30B-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf unsloth/Muse-Glimmer-30B-GGUF:UD-Q4_K_XL

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "unsloth/Muse-Glimmer-30B-GGUF:UD-Q4_K_XL" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

llama Error

by rnr1 - opened 11 days ago

Discussion

rnr1

11 days ago

llama.cpp throws error:
llama_model_load: error loading model: unknown model architecture: 'muse-glimmer'

graal072

11 days ago

same

emircansevdi

11 days ago

Update your Llama.cpp software. Updating by compiling from source code ensures you have the latest version.

deuxenun

11 days ago

As of writing this message support was merged 1hr ago https://github.com/ggml-org/llama.cpp/pull/26841

Recompile yourself or wait for a couple of hours until it hits official releases

SergioPatricio

11 days ago

Latest build b10342 print:

Loading model... \0.00.269.162 E llama_model_load: error loading model: unknown model architecture: 'muse-glimmer'
0.00.269.165 E llama_model_load_from_file_impl: failed to load model
0.00.269.186 E common_fit_params: encountered an error while trying to fit params to free device memory: failed to load model
|0.00.347.567 E llama_model_load: error loading model: unknown model architecture: 'muse-glimmer'
0.00.347.569 E llama_model_load_from_file_impl: failed to load model
0.00.347.572 E cmn common_init_: failed to load model '/data/AI/llama.cpp/models/Muse-Glimmer-30B-UD-IQ2_XXS.gguf'
0.00.347.573 E srv load_model: failed to load model, '/data/AI/llama.cpp/models/Muse-Glimmer-30B-UD-IQ2_XXS.gguf'
llama_server exited with code 1
0.00.350.316 E srv llama_server: exiting due to model loading error

pegasus912

11 days ago

Same here, on the latest build for Vulkan.

MaxDaddyLongs

11 days ago

same

pegasus912

11 days ago

It does seem to be isolated to Vulkan from my testing. It works with Lemonade's latest ROCm build of llama.cpp

Mikeczyk

11 days ago

With latest llama.cpp build I got Vulkan backend to work with split mode layer, with split mode tensor it errors out building the context for the dflash model. This is on Nvidia GPUs but that shouldn't matter.

Split mode layer

root@llama:/opt/llama.cpp# ./llama-server-vulkan --device Vulkan0,Vulkan1 \
-m "/AI/Muse-Glimmer-30B-UD-Q8_K_XL.gguf" \
--spec-type draft-dflash --spec-draft-p-min 0.2 --spec-draft-n-min 0 --spec-draft-n-max 3 \
--parallel 2 --ctx-size 262144 --batch-size 2048 --ubatch-size 1024 \
--jinja --host 0.0.0.0 --port 8080 --spec-draft-model /AI/dflash-kquant.gguf \
--cache-type-k q8_0 --cache-type-v q8_0 -ctvd q8_0 -ctkd q8_0 --mmproj /AI/mmproj-Muse-Glimmer-30B-Q8_0.gguf
0.00.581.142 I cmn  common_param: common_params_print_info: verbosity = 3 (adjust with the `-lv N` CLI arg)
0.00.588.027 W srv  llama_server: -----------------
0.00.588.030 W srv  llama_server: CORS is set to allow all origins ('*') and no API key is set
0.00.588.031 W srv  llama_server: this can be a security risk (cross-origin attacks)
0.00.588.031 W srv  llama_server: more info: https://github.com/ggml-org/llama.cpp/pull/25655
0.00.588.031 W srv  llama_server: -----------------
0.00.589.402 I srv    load_model: loading model '/AI/Muse-Glimmer-30B-UD-Q8_K_XL.gguf'
0.02.519.160 E llama_init_from_model: failed to initialize the context: dflash requires ctx_other to be set (this warning is normal during memory fitting)
0.02.633.712 W srv    load_model: [spec] failed to measure draft model memory: failed to create llama_context from model
0.04.246.668 W load: setting token '<|message|>' (200023) attribute to USER_DEFINED (16), old attributes: 8
0.04.246.674 W load: setting token '<|start|>' (200022) attribute to USER_DEFINED (16), old attributes: 8
0.04.263.666 W load: special_eot_id is not in special_eog_ids - the tokenizer config may be incorrect
2.41.682.655 I common_speculative_init_result: loading draft model '/AI/dflash-kquant.gguf'
2.42.458.263 W load: setting token '<|message|>' (200023) attribute to USER_DEFINED (16), old attributes: 8
2.42.458.270 W load: setting token '<|start|>' (200022) attribute to USER_DEFINED (16), old attributes: 8
2.42.477.900 W load: special_eot_id is not in special_eog_ids - the tokenizer config may be incorrect
3.01.974.215 I srv    load_model: loaded multimodal model, '/AI/mmproj-Muse-Glimmer-30B-Q8_0.gguf'
3.02.570.391 I srv    load_model: initializing, n_slots = 2, n_ctx_slot = 131072, kv_unified = 'false'
3.02.571.356 I common_speculative_impl_draft_dflash: adding speculative implementation 'draft-dflash'
3.02.571.368 I common_speculative_impl_draft_dflash: - n_max=3, n_min=0, p_min=0.20
3.02.571.369 I common_speculative_impl_draft_dflash: - block_size=16, mask_token_id=201818, n_extract=5
3.03.480.629 I srv          init: chat template supports preserving reasoning, consider enabling it via --reasoning-preserve
3.03.480.858 I srv  llama_server: model loaded
3.03.480.864 I srv  llama_server: listening on http://0.0.0.0:8080

Split mode tensor

root@llama:/opt/llama.cpp# ./llama-server-vulkan --device Vulkan2,Vulkan3 \
-m "/AI/Muse-Glimmer-30B-UD-Q8_K_XL.gguf" --spec-type draft-dflash \
--spec-draft-p-min 0.2 --spec-draft-n-min 0 --spec-draft-n-max 3 \
--parallel 2 --ctx-size 262144 --batch-size 2048 --ubatch-size 1024 --jinja \
--host 0.0.0.0 --port 8080 --spec-draft-model /AI/dflash-kquant.gguf -sm tensor \
--cache-type-k q8_0 --cache-type-v q8_0 -ctvd q8_0 -ctkd q8_0 \
--mmproj /AI/mmproj-Muse-Glimmer-30B-Q8_0.gguf
0.00.334.560 I cmn  common_param: common_params_print_info: verbosity = 3 (adjust with the `-lv N` CLI arg)
0.00.340.819 W srv  llama_server: -----------------
0.00.340.823 W srv  llama_server: CORS is set to allow all origins ('*') and no API key is set
0.00.340.823 W srv  llama_server: this can be a security risk (cross-origin attacks)
0.00.340.824 W srv  llama_server: more info: https://github.com/ggml-org/llama.cpp/pull/25655
0.00.340.824 W srv  llama_server: -----------------
0.00.342.198 I srv    load_model: loading model '/AI/Muse-Glimmer-30B-UD-Q8_K_XL.gguf'
0.02.224.064 E llama_init_from_model: failed to initialize the context: dflash requires ctx_other to be set (this warning is normal during memory fitting)
0.02.341.526 W srv    load_model: [spec] failed to measure draft model memory: failed to create llama_context from model
0.02.341.567 W common_fit_params: failed to fit params to free device memory: llama_params_fit is not implemented for SPLIT_MODE_TENSOR, abort
0.03.020.761 W load: setting token '<|message|>' (200023) attribute to USER_DEFINED (16), old attributes: 8
0.03.020.771 W load: setting token '<|start|>' (200022) attribute to USER_DEFINED (16), old attributes: 8
0.03.038.832 W load: special_eot_id is not in special_eog_ids - the tokenizer config may be incorrect
4.07.136.238 I common_speculative_init_result: loading draft model '/AI/dflash-kquant.gguf'
4.07.868.161 W load: setting token '<|message|>' (200023) attribute to USER_DEFINED (16), old attributes: 8
4.07.868.166 W load: setting token '<|start|>' (200022) attribute to USER_DEFINED (16), old attributes: 8
4.07.886.881 W load: special_eot_id is not in special_eog_ids - the tokenizer config may be incorrect
/opt/llama.cpp/ggml/src/ggml-backend.cpp:898: pre-allocated tensor (output.weight) in a buffer (Meta()) that cannot run the operation (NONE)
/opt/llama.cpp/vulkan/libggml-base.so.0(+0x1a058) [0x7a61b5256058]
/opt/llama.cpp/vulkan/libggml-base.so.0(ggml_print_backtrace+0x1e4) [0x7a61b5256434]
/opt/llama.cpp/vulkan/libggml-base.so.0(ggml_abort+0x11e) [0x7a61b52565be]
/opt/llama.cpp/vulkan/libggml-base.so.0(+0x32d5c) [0x7a61b526ed5c]
/opt/llama.cpp/vulkan/libggml-base.so.0(ggml_backend_sched_split_graph+0xc9c) [0x7a61b5270bdc]
/opt/llama.cpp/vulkan/libllama.so.0(_ZN13llama_context13graph_reserveEjjjPK22llama_memory_context_ibPm+0x629) [0x7a61b3ce7799]
/opt/llama.cpp/vulkan/libllama.so.0(_ZN13llama_context17resolve_fused_opsEPK22llama_memory_context_ij+0x142) [0x7a61b3ce7bc2]
/opt/llama.cpp/vulkan/libllama.so.0(_ZN13llama_context13sched_reserveEv+0x634) [0x7a61b3ce8ba4]
/opt/llama.cpp/vulkan/libllama.so.0(_ZN13llama_contextC1ERK11llama_model20llama_context_params+0xcda) [0x7a61b3cec8aa]
/opt/llama.cpp/vulkan/libllama.so.0(llama_init_from_model+0x2e9) [0x7a61b3cedbc9]
/opt/llama.cpp/vulkan/libllama-common.so.0(_ZN30common_speculative_init_resultC1ER13common_paramsP11llama_modelP13llama_context+0x449) [0x7a61b42d4be9]
/opt/llama.cpp/vulkan/libllama-common.so.0(_Z35common_speculative_init_from_paramsR13common_paramsP11llama_modelP13llama_context+0x32) [0x7a61b42d4ec2]
/opt/llama.cpp/vulkan/libllama-server-impl.so(_ZN19server_context_impl10load_modelER13common_params+0x1c89) [0x7a61b4db8ff9]
/opt/llama.cpp/vulkan/libllama-server-impl.so(_Z12llama_serverR13common_paramsiPPc+0x40b3) [0x7a61b4cf1443]
/opt/llama.cpp/vulkan/libllama-server-impl.so(_Z12llama_serveriPPc+0x9e) [0x7a61b4cf26ee]
/lib/x86_64-linux-gnu/libc.so.6(+0x29ca8) [0x7a61b4635ca8]
/lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0x85) [0x7a61b4635d65]
./llama-server-vulkan(+0x11b1) [0x58d0921af1b1]
Aborted

Split mode tensor cuda

root@llama:/opt/llama.cpp# CUDA_VISIBLE_DEVICES=2,3 \
./llama-server -m "/AI/Muse-Glimmer-30B-UD-Q8_K_XL.gguf" \
--spec-type draft-dflash --spec-draft-p-min 0.2 --spec-draft-n-min 0 --spec-draft-n-max 3 \
--parallel 2 --ctx-size 262144 --batch-size 2048 --ubatch-size 1024 --jinja \
--host 0.0.0.0 --port 8080 --spec-draft-model /AI/dflash-kquant.gguf -sm tensor \
--cache-type-k q8_0 --cache-type-v q8_0 -ctvd q8_0 -ctkd q8_0 --mmproj /AI/mmproj-Muse-Glimmer-30B-Q8_0.gguf
0.00.074.572 I cmn  common_param: common_params_print_info: verbosity = 3 (adjust with the `-lv N` CLI arg)
0.00.386.519 W srv  llama_server: -----------------
0.00.386.522 W srv  llama_server: CORS is set to allow all origins ('*') and no API key is set
0.00.386.523 W srv  llama_server: this can be a security risk (cross-origin attacks)
0.00.386.523 W srv  llama_server: more info: https://github.com/ggml-org/llama.cpp/pull/25655
0.00.386.523 W srv  llama_server: -----------------
0.00.388.768 I srv    load_model: loading model '/AI/Muse-Glimmer-30B-UD-Q8_K_XL.gguf'
0.01.339.315 E llama_init_from_model: failed to initialize the context: dflash requires ctx_other to be set (this warning is normal during memory fitting)
0.01.463.248 W srv    load_model: [spec] failed to measure draft model memory: failed to create llama_context from model
0.01.463.293 W common_fit_params: failed to fit params to free device memory: llama_params_fit is not implemented for SPLIT_MODE_TENSOR, abort
0.02.135.342 W load: setting token '<|message|>' (200023) attribute to USER_DEFINED (16), old attributes: 8
0.02.135.350 W load: setting token '<|start|>' (200022) attribute to USER_DEFINED (16), old attributes: 8
0.02.152.465 W load: special_eot_id is not in special_eog_ids - the tokenizer config may be incorrect
3.32.079.551 I common_speculative_init_result: loading draft model '/AI/dflash-kquant.gguf'
3.32.831.298 W load: setting token '<|message|>' (200023) attribute to USER_DEFINED (16), old attributes: 8
3.32.831.305 W load: setting token '<|start|>' (200022) attribute to USER_DEFINED (16), old attributes: 8
3.32.848.376 W load: special_eot_id is not in special_eog_ids - the tokenizer config may be incorrect
3.56.344.218 I srv    load_model: loaded multimodal model, '/AI/mmproj-Muse-Glimmer-30B-Q8_0.gguf'
3.56.729.025 I srv    load_model: initializing, n_slots = 2, n_ctx_slot = 131072, kv_unified = 'false'
3.56.729.984 I common_speculative_impl_draft_dflash: adding speculative implementation 'draft-dflash'
3.56.729.989 I common_speculative_impl_draft_dflash: - n_max=3, n_min=0, p_min=0.20
3.56.729.990 I common_speculative_impl_draft_dflash: - block_size=16, mask_token_id=201818, n_extract=5
3.56.845.186 I srv          init: chat template supports preserving reasoning, consider enabling it via --reasoning-preserve
3.56.845.430 I srv  llama_server: model loaded
3.56.845.436 I srv  llama_server: listening on http://0.0.0.0:8080

seagullowl

10 days ago

•

edited 10 days ago

server-vulkan-b10362 can run Muse Glimmer now.

rahul7star

7 days ago

2 core cpu 16gb ram

0.00.082.434 I srv    load_model: loading model '/app/Muse-Glimmer-30B-UD-Q2_K_XL.gguf'
0.01.755.297 W load: setting token '<|message|>' (200023) attribute to USER_DEFINED (16), old attributes: 8
0.01.755.307 W load: setting token '<|start|>' (200022) attribute to USER_DEFINED (16), old attributes: 8
0.01.771.098 W load: special_eot_id is not in special_eog_ids - the tokenizer config may be incorrect
0.31.965.115 I srv    load_model: loaded multimodal model, '/app/mmproj-Muse-Glimmer-30B-BF16.gguf'
0.34.104.328 I srv    load_model: initializing, n_slots = 4, n_ctx_slot = 64000, kv_unified = 'true'
0.34.143.256 I srv          init: chat template supports preserving reasoning, consider enabling it via --reasoning-preserve
0.34.143.341 I srv  llama_server: model loaded
0.34.143.351 I srv  llama_server: listening on http://0.0.0.0:7860
1.25.506.834 I slot get_availabl: id  3 | task -1 | selected slot by LRU, t_last = -1

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment