Instructions to use yeasah/Muse-Glimmer-30B-exl3 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use yeasah/Muse-Glimmer-30B-exl3 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-text-to-text", model="yeasah/Muse-Glimmer-30B-exl3")# pip install -U transformers accelerate # Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("yeasah/Muse-Glimmer-30B-exl3", device_map="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use yeasah/Muse-Glimmer-30B-exl3 with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "yeasah/Muse-Glimmer-30B-exl3" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "yeasah/Muse-Glimmer-30B-exl3", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/yeasah/Muse-Glimmer-30B-exl3
- SGLang
How to use yeasah/Muse-Glimmer-30B-exl3 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 "yeasah/Muse-Glimmer-30B-exl3" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "yeasah/Muse-Glimmer-30B-exl3", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'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 "yeasah/Muse-Glimmer-30B-exl3" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "yeasah/Muse-Glimmer-30B-exl3", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use yeasah/Muse-Glimmer-30B-exl3 with Docker Model Runner:
docker model run hf.co/yeasah/Muse-Glimmer-30B-exl3
Download qb_self.json from yeasah/Muse-Glimmer-30B-exl3: direct link, hf CLI and curl.
- Browser
- Download file 12.4 kB
-
https://huggingface.co/yeasah/Muse-Glimmer-30B-exl3/resolve/main/qb_self.json
- Command line
-
hf download hf://yeasah/Muse-Glimmer-30B-exl3/qb_self.json
-
curl -L -o qb_self.json https://huggingface.co/yeasah/Muse-Glimmer-30B-exl3/resolve/main/qb_self.json
12.4 kB
| [ | |
| { | |
| "label": "HF BF16", | |
| "group": "reference", | |
| "ppl": 1.5350952833692508, | |
| "bpw_layer": 16.0, | |
| "bpw_head": 16.0, | |
| "bpw_embed": 16.0, | |
| "vram_gb": 51.8809814453125 | |
| }, | |
| { | |
| "label": "Noise floor", | |
| "group": "noise_floor", | |
| "ppl": 1.535574427165267, | |
| "kld": 0.0009139982284978032, | |
| "kld_median": 4.605692811310291e-05, | |
| "kld_p10": -8.451653599195197e-09, | |
| "kld_p25": 7.736846185935065e-08, | |
| "kld_p75": 0.0008647915674373507, | |
| "kld_p90": 0.0023589718621224165, | |
| "kld_buckets": [ | |
| { | |
| "conf_range": [ | |
| 0.0, | |
| 0.25 | |
| ], | |
| "share": 0.015772243961691856, | |
| "mean": 0.002427938161417842, | |
| "median": 0.0017011184245347977 | |
| }, | |
| { | |
| "conf_range": [ | |
| 0.25, | |
| 0.5 | |
| ], | |
| "share": 0.11298482120037079, | |
| "mean": 0.002774390857666731, | |
| "median": 0.0017480563838034868 | |
| }, | |
| { | |
| "conf_range": [ | |
| 0.5, | |
| 0.75 | |
| ], | |
| "share": 0.162855863571167, | |
| "mean": 0.0021461709402501583, | |
| "median": 0.0012395614758133888 | |
| }, | |
| { | |
| "conf_range": [ | |
| 0.75, | |
| 0.95 | |
| ], | |
| "share": 0.16585655510425568, | |
| "mean": 0.0011193135287612677, | |
| "median": 0.0004438331816345453 | |
| }, | |
| { | |
| "conf_range": [ | |
| 0.95, | |
| 1.0 | |
| ], | |
| "share": 0.5425304770469666, | |
| "mean": 4.991054447600618e-05, | |
| "median": 2.3558092721032153e-07 | |
| } | |
| ], | |
| "bpw_layer": 16.0, | |
| "bpw_head": 16.0, | |
| "bpw_embed": 16.0, | |
| "vram_gb": 51.8809814453125 | |
| }, | |
| { | |
| "label": "2.50bpw-H5", | |
| "group": "EXL3", | |
| "ppl": 1.755496804107692, | |
| "kld": 0.12456365674734116, | |
| "kld_median": 0.017651762813329697, | |
| "kld_p10": 9.3828077751823e-08, | |
| "kld_p25": 6.163350190035999e-05, | |
| "kld_p75": 0.11416993290185928, | |
| "kld_p90": 0.3378671109676361, | |
| "kld_buckets": [ | |
| { | |
| "conf_range": [ | |
| 0.0, | |
| 0.25 | |
| ], | |
| "share": 0.015772243961691856, | |
| "mean": 0.18763397634029388, | |
| "median": 0.05495230108499527 | |
| }, | |
| { | |
| "conf_range": [ | |
| 0.25, | |
| 0.5 | |
| ], | |
| "share": 0.11298482120037079, | |
| "mean": 0.27698227763175964, | |
| "median": 0.15223616361618042 | |
| }, | |
| { | |
| "conf_range": [ | |
| 0.5, | |
| 0.75 | |
| ], | |
| "share": 0.162855863571167, | |
| "mean": 0.25408294796943665, | |
| "median": 0.12243825197219849 | |
| }, | |
| { | |
| "conf_range": [ | |
| 0.75, | |
| 0.95 | |
| ], | |
| "share": 0.16585655510425568, | |
| "mean": 0.199837327003479, | |
| "median": 0.0798921138048172 | |
| }, | |
| { | |
| "conf_range": [ | |
| 0.95, | |
| 1.0 | |
| ], | |
| "share": 0.5425304770469666, | |
| "mean": 0.029097381979227066, | |
| "median": 0.00012891614460386336 | |
| } | |
| ], | |
| "bpw_layer": 2.504164938549567, | |
| "bpw_head": 5.00248303396898, | |
| "bpw_embed": 16.0, | |
| "vram_gb": 10.624184969812632, | |
| "on_disk_gb": 14.20649278536439, | |
| "unexplained_gb": 0.0, | |
| "expected_absent_gb": 3.582307815551758 | |
| }, | |
| { | |
| "label": "3.00bpw-H5", | |
| "group": "EXL3", | |
| "ppl": 1.6380082887788623, | |
| "kld": 0.06327185034751892, | |
| "kld_median": 0.00632548937574029, | |
| "kld_p10": 5.393940155640209e-11, | |
| "kld_p25": 1.2382403838273603e-05, | |
| "kld_p75": 0.05211612954735756, | |
| "kld_p90": 0.1702725887298584, | |
| "kld_buckets": [ | |
| { | |
| "conf_range": [ | |
| 0.0, | |
| 0.25 | |
| ], | |
| "share": 0.015772243961691856, | |
| "mean": 0.09361827373504639, | |
| "median": 0.032146818935871124 | |
| }, | |
| { | |
| "conf_range": [ | |
| 0.25, | |
| 0.5 | |
| ], | |
| "share": 0.11298482120037079, | |
| "mean": 0.15589560568332672, | |
| "median": 0.07994944602251053 | |
| }, | |
| { | |
| "conf_range": [ | |
| 0.5, | |
| 0.75 | |
| ], | |
| "share": 0.162855863571167, | |
| "mean": 0.13927622139453888, | |
| "median": 0.06277595460414886 | |
| }, | |
| { | |
| "conf_range": [ | |
| 0.75, | |
| 0.95 | |
| ], | |
| "share": 0.16585655510425568, | |
| "mean": 0.09841413050889969, | |
| "median": 0.03573772311210632 | |
| }, | |
| { | |
| "conf_range": [ | |
| 0.95, | |
| 1.0 | |
| ], | |
| "share": 0.5425304770469666, | |
| "mean": 0.009542074054479599, | |
| "median": 2.675077303138096e-05 | |
| } | |
| ], | |
| "bpw_layer": 3.004164938549567, | |
| "bpw_head": 5.00248303396898, | |
| "bpw_embed": 16.0, | |
| "vram_gb": 12.088906649500132, | |
| "on_disk_gb": 15.67121446505189, | |
| "unexplained_gb": 0.0, | |
| "expected_absent_gb": 3.582307815551758 | |
| }, | |
| { | |
| "label": "3.50bpw-H5", | |
| "group": "EXL3", | |
| "ppl": 1.5876834559549282, | |
| "kld": 0.029775694012641907, | |
| "kld_median": 0.0026424750685691833, | |
| "kld_p10": -4.3649617254004625e-10, | |
| "kld_p25": 4.441782039066311e-06, | |
| "kld_p75": 0.022654403001070023, | |
| "kld_p90": 0.0781114473938942, | |
| "kld_buckets": [ | |
| { | |
| "conf_range": [ | |
| 0.0, | |
| 0.25 | |
| ], | |
| "share": 0.015772243961691856, | |
| "mean": 0.047659460455179214, | |
| "median": 0.013781007379293442 | |
| }, | |
| { | |
| "conf_range": [ | |
| 0.25, | |
| 0.5 | |
| ], | |
| "share": 0.11298482120037079, | |
| "mean": 0.0798071101307869, | |
| "median": 0.036406341940164566 | |
| }, | |
| { | |
| "conf_range": [ | |
| 0.5, | |
| 0.75 | |
| ], | |
| "share": 0.162855863571167, | |
| "mean": 0.06722264736890793, | |
| "median": 0.029358290135860443 | |
| }, | |
| { | |
| "conf_range": [ | |
| 0.75, | |
| 0.95 | |
| ], | |
| "share": 0.16585655510425568, | |
| "mean": 0.043947238475084305, | |
| "median": 0.015285729430615902 | |
| }, | |
| { | |
| "conf_range": [ | |
| 0.95, | |
| 1.0 | |
| ], | |
| "share": 0.5425304770469666, | |
| "mean": 0.003263351507484913, | |
| "median": 9.27980181586463e-06 | |
| } | |
| ], | |
| "bpw_layer": 3.504164938549567, | |
| "bpw_head": 5.00248303396898, | |
| "bpw_embed": 16.0, | |
| "vram_gb": 13.553628329187632, | |
| "on_disk_gb": 17.13593614473939, | |
| "unexplained_gb": 0.0, | |
| "expected_absent_gb": 3.582307815551758 | |
| }, | |
| { | |
| "label": "4.00bpw", | |
| "group": "EXL3", | |
| "ppl": 1.561660829458369, | |
| "kld": 0.016509652137756348, | |
| "kld_median": 0.0012908224016427994, | |
| "kld_p10": -2.3589903364040765e-09, | |
| "kld_p25": 1.7974593902181368e-06, | |
| "kld_p75": 0.01253344863653183, | |
| "kld_p90": 0.04319066181778908, | |
| "kld_buckets": [ | |
| { | |
| "conf_range": [ | |
| 0.0, | |
| 0.25 | |
| ], | |
| "share": 0.015772243961691856, | |
| "mean": 0.02720433659851551, | |
| "median": 0.007815390825271606 | |
| }, | |
| { | |
| "conf_range": [ | |
| 0.25, | |
| 0.5 | |
| ], | |
| "share": 0.11298482120037079, | |
| "mean": 0.044597309082746506, | |
| "median": 0.021441051736474037 | |
| }, | |
| { | |
| "conf_range": [ | |
| 0.5, | |
| 0.75 | |
| ], | |
| "share": 0.162855863571167, | |
| "mean": 0.03968896344304085, | |
| "median": 0.01667233183979988 | |
| }, | |
| { | |
| "conf_range": [ | |
| 0.75, | |
| 0.95 | |
| ], | |
| "share": 0.16585655510425568, | |
| "mean": 0.022727709263563156, | |
| "median": 0.008285239338874817 | |
| }, | |
| { | |
| "conf_range": [ | |
| 0.95, | |
| 1.0 | |
| ], | |
| "share": 0.5425304770469666, | |
| "mean": 0.0014904960989952087, | |
| "median": 3.590653022911283e-06 | |
| } | |
| ], | |
| "bpw_layer": 4.004164938549566, | |
| "bpw_head": 6.00248303396898, | |
| "bpw_embed": 16.0, | |
| "vram_gb": 15.1749585904181, | |
| "on_disk_gb": 18.757266405969858, | |
| "unexplained_gb": 0.0, | |
| "expected_absent_gb": 3.582307815551758 | |
| }, | |
| { | |
| "label": "5.00bpw", | |
| "group": "EXL3", | |
| "ppl": 1.540988700428853, | |
| "kld": 0.004771005362272263, | |
| "kld_median": 0.00037156976759433746, | |
| "kld_p10": -2.0434434144789293e-09, | |
| "kld_p25": 6.524813329633616e-07, | |
| "kld_p75": 0.0037913480773568153, | |
| "kld_p90": 0.012503254227340221, | |
| "kld_buckets": [ | |
| { | |
| "conf_range": [ | |
| 0.0, | |
| 0.25 | |
| ], | |
| "share": 0.015772243961691856, | |
| "mean": 0.008537128567695618, | |
| "median": 0.0031840484589338303 | |
| }, | |
| { | |
| "conf_range": [ | |
| 0.25, | |
| 0.5 | |
| ], | |
| "share": 0.11298482120037079, | |
| "mean": 0.013262076303362846, | |
| "median": 0.0066865552216768265 | |
| }, | |
| { | |
| "conf_range": [ | |
| 0.5, | |
| 0.75 | |
| ], | |
| "share": 0.162855863571167, | |
| "mean": 0.011689701117575169, | |
| "median": 0.005038066767156124 | |
| }, | |
| { | |
| "conf_range": [ | |
| 0.75, | |
| 0.95 | |
| ], | |
| "share": 0.16585655510425568, | |
| "mean": 0.006397041492164135, | |
| "median": 0.002399744000285864 | |
| }, | |
| { | |
| "conf_range": [ | |
| 0.95, | |
| 1.0 | |
| ], | |
| "share": 0.5425304770469666, | |
| "mean": 0.0003192703006789088, | |
| "median": 1.1232980341446819e-06 | |
| } | |
| ], | |
| "bpw_layer": 5.004164938549566, | |
| "bpw_head": 6.00248303396898, | |
| "bpw_embed": 16.0, | |
| "vram_gb": 18.1044019497931, | |
| "on_disk_gb": 21.686709765344858, | |
| "unexplained_gb": 0.0, | |
| "expected_absent_gb": 3.582307815551758 | |
| }, | |
| { | |
| "label": "6.00bpw", | |
| "group": "EXL3", | |
| "ppl": 1.536404394752159, | |
| "kld": 0.0017247830983251333, | |
| "kld_median": 0.0001451806165277958, | |
| "kld_p10": -4.444443035822587e-09, | |
| "kld_p25": 3.0863679967296775e-07, | |
| "kld_p75": 0.0015155579894781113, | |
| "kld_p90": 0.004657591227442026, | |
| "kld_buckets": [ | |
| { | |
| "conf_range": [ | |
| 0.0, | |
| 0.25 | |
| ], | |
| "share": 0.015772243961691856, | |
| "mean": 0.003872393863275647, | |
| "median": 0.0018864693120121956 | |
| }, | |
| { | |
| "conf_range": [ | |
| 0.25, | |
| 0.5 | |
| ], | |
| "share": 0.11298482120037079, | |
| "mean": 0.005033925641328096, | |
| "median": 0.00273834727704525 | |
| }, | |
| { | |
| "conf_range": [ | |
| 0.5, | |
| 0.75 | |
| ], | |
| "share": 0.162855863571167, | |
| "mean": 0.004184214398264885, | |
| "median": 0.001957982312887907 | |
| }, | |
| { | |
| "conf_range": [ | |
| 0.75, | |
| 0.95 | |
| ], | |
| "share": 0.16585655510425568, | |
| "mean": 0.002147519262507558, | |
| "median": 0.0008811890729703009 | |
| }, | |
| { | |
| "conf_range": [ | |
| 0.95, | |
| 1.0 | |
| ], | |
| "share": 0.5425304770469666, | |
| "mean": 0.00010569998994469643, | |
| "median": 5.994631919747917e-07 | |
| } | |
| ], | |
| "bpw_layer": 6.004164938549566, | |
| "bpw_head": 6.00248303396898, | |
| "bpw_embed": 16.0, | |
| "vram_gb": 21.0338453091681, | |
| "on_disk_gb": 24.616153124719858, | |
| "unexplained_gb": 0.0, | |
| "expected_absent_gb": 3.582307815551758 | |
| }, | |
| { | |
| "label": "thr:RedHatAI/Muse-Glimmer-30B-FP8-block", | |
| "group": "threshold", | |
| "ppl": 1.5436832119692534, | |
| "kld": 0.006228696554899216, | |
| "kld_median": 0.0004936084151268005, | |
| "kld_p10": -3.2378028036461615e-10, | |
| "kld_p25": 7.255482614709763e-07, | |
| "kld_p75": 0.005493507720530033, | |
| "kld_p90": 0.016667133197188377, | |
| "kld_buckets": [ | |
| { | |
| "conf_range": [ | |
| 0.0, | |
| 0.25 | |
| ], | |
| "share": 0.015772243961691856, | |
| "mean": 0.012596595101058483, | |
| "median": 0.007302751764655113 | |
| }, | |
| { | |
| "conf_range": [ | |
| 0.25, | |
| 0.5 | |
| ], | |
| "share": 0.11298482120037079, | |
| "mean": 0.018517345190048218, | |
| "median": 0.010037282481789589 | |
| }, | |
| { | |
| "conf_range": [ | |
| 0.5, | |
| 0.75 | |
| ], | |
| "share": 0.162855863571167, | |
| "mean": 0.014766950160264969, | |
| "median": 0.007181471213698387 | |
| }, | |
| { | |
| "conf_range": [ | |
| 0.75, | |
| 0.95 | |
| ], | |
| "share": 0.16585655510425568, | |
| "mean": 0.007918265648186207, | |
| "median": 0.003250366309657693 | |
| }, | |
| { | |
| "conf_range": [ | |
| 0.95, | |
| 1.0 | |
| ], | |
| "share": 0.5425304770469666, | |
| "mean": 0.00040488041122443974, | |
| "median": 1.2951336429978255e-06 | |
| } | |
| ], | |
| "bpw_layer": 8.001953125, | |
| "bpw_head": 16.0, | |
| "bpw_embed": 16.0, | |
| "vram_gb": 28.45115613937378 | |
| } | |
| ] |