electron-rare/exo
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

clean model cards

Browse Source
This commit is contained in:
Alex Cheema
2025-08-04 05:31:30 +08:00
parent f352ddfc5f
commit 6daf7f31f7
1 changed files with 79 additions and 100 deletions
Download Patch File Download Diff File Expand all files Collapse all files
shared/models/model_cards.py
+79 -100
View File
@@ -15,86 +15,66 @@ class ModelCard(BaseModel):
MODEL_CARDS: dict[str, ModelCard] = {
"deepseek-v3-0324": ModelCard(
# deepseek v3
"deepseek-v3-0324:4bit": ModelCard(
short_id="deepseek-v3-0324",
model_id="mlx-community/DeepSeek-v3-0324-8bit",
name="DeepSeek V3 fp8",
model_id="mlx-community/DeepSeek-V3-0324-4bit",
name="DeepSeek V3 0324 (4-bit)",
description="""DeepSeek V3 is a large language model trained on the DeepSeek V3 dataset.""",
tags=[],
metadata=ModelMetadata(
model_id="mlx-community/DeepSeek-v3-0324-8bit",
pretty_name="DeepSeek V3 fp8",
model_id="mlx-community/DeepSeek-V3-0324-4bit",
pretty_name="DeepSeek V3 0324 (4-bit)",
storage_size_kilobytes=754998771712//1024,
n_layers=61,
),
),
"llama-3.3": ModelCard(
short_id="llama-3.3",
model_id="mlx-community/Llama-3.3-70B-Instruct-4bit",
name="Llama 3.3 70B",
description="""The Meta Llama 3.3 multilingual large language model (LLM) is an instruction tuned generative model in 70B (text in/text out)""",
"deepseek-v3-0324": ModelCard(
short_id="deepseek-v3-0324",
model_id="mlx-community/DeepSeek-v3-0324-8bit",
name="DeepSeek V3 0324 (8-bit)",
description="""DeepSeek V3 is a large language model trained on the DeepSeek V3 dataset.""",
tags=[],
metadata=ModelMetadata(
model_id="mlx-community/Llama-3.3-70B-Instruct-4bit",
pretty_name="Llama 3.3 70B",
storage_size_kilobytes=38758160,
n_layers=80,
model_id="mlx-community/DeepSeek-v3-0324-8bit",
pretty_name="DeepSeek V3 0324 (8-bit)",
storage_size_kilobytes=754998771712//1024,
n_layers=61,
),
),
"llama-3.3:70b": ModelCard(
short_id="llama-3.3:70b",
model_id="mlx-community/Llama-3.3-70B-Instruct-4bit",
name="Llama 3.3 70B",
description="""The Meta Llama 3.3 multilingual large language model (LLM) is an instruction tuned generative model in 70B (text in/text out)""",
# deepseek r1
"deepseek-r1-0528:4bit": ModelCard(
short_id="deepseek-r1-0528",
model_id="mlx-community/DeepSeek-R1-0528-4bit",
name="DeepSeek-R1-0528 (4-bit)",
description="""DeepSeek R1 is a large language model trained on the DeepSeek R1 dataset.""",
tags=[],
metadata=ModelMetadata(
model_id="mlx-community/Llama-3.3-70B-Instruct-4bit",
pretty_name="Llama 3.3 70B",
storage_size_kilobytes=38758160,
n_layers=80,
model_id="mlx-community/DeepSeek-R1-0528-4bit",
pretty_name="DeepSeek R1 671B (4-bit)",
storage_size_kilobytes=409706307,
n_layers=61,
),
),
"llama-3.2": ModelCard(
short_id="llama-3.2",
model_id="mlx-community/Llama-3.2-1B-Instruct-4bit",
name="Llama 3.2 1B",
description="""Llama 3.2 is a large language model trained on the Llama 3.2 dataset.""",
"deepseek-r1-0528": ModelCard(
short_id="deepseek-r1-0528",
model_id="mlx-community/DeepSeek-R1-0528-8bit",
name="DeepSeek-R1-0528 (8-bit)",
description="""DeepSeek R1 is a large language model trained on the DeepSeek R1 dataset.""",
tags=[],
metadata=ModelMetadata(
model_id="mlx-community/Llama-3.2-1B-Instruct-4bit",
pretty_name="Llama 3.2 1B",
storage_size_kilobytes=678948,
n_layers=16,
model_id="mlx-community/DeepSeek-R1-0528-8bit",
pretty_name="DeepSeek R1 671B (8-bit)",
storage_size_kilobytes=409706307,
n_layers=61,
),
),
"llama-3.2:1b": ModelCard(
short_id="llama-3.2:1b",
model_id="mlx-community/Llama-3.2-1B-Instruct-4bit",
name="Llama 3.2 1B",
description="""Llama 3.2 is a large language model trained on the Llama 3.2 dataset.""",
tags=[],
metadata=ModelMetadata(
model_id="mlx-community/Llama-3.2-1B-Instruct-4bit",
pretty_name="Llama 3.2 1B",
storage_size_kilobytes=678948,
n_layers=16,
),
),
"llama-3.2:3b": ModelCard(
short_id="llama-3.2:3b",
model_id="mlx-community/Llama-3.2-3B-Instruct-4bit",
name="Llama 3.2 3B",
description="""Llama 3.2 is a large language model trained on the Llama 3.2 dataset.""",
tags=[],
metadata=ModelMetadata(
model_id="mlx-community/Llama-3.2-3B-Instruct-4bit",
pretty_name="Llama 3.2 3B",
storage_size_kilobytes=1765062,
n_layers=28,
),
),
"llama-3.1:8b": ModelCard(
short_id="llama-3.1:8b",
# llama-3.1
"llama-3.1-8b": ModelCard(
short_id="llama-3.1-8b",
model_id="mlx-community/Meta-Llama-3.1-8B-Instruct-4bit",
name="Llama 3.1 8B",
description="""Llama 3.1 is a large language model trained on the Llama 3.1 dataset.""",
@@ -119,58 +99,51 @@ MODEL_CARDS: dict[str, ModelCard] = {
n_layers=80,
),
),
"deepseek-r1": ModelCard(
short_id="deepseek-r1",
model_id="mlx-community/DeepSeek-R1-4bit",
name="DeepSeek R1 671B (4-bit)",
description="""DeepSeek R1 is a large language model trained on the DeepSeek R1 dataset.""",
# llama-3.2
"llama-3.2-1b": ModelCard(
short_id="llama-3.2-1b",
model_id="mlx-community/Llama-3.2-1B-Instruct-4bit",
name="Llama 3.2 1B",
description="""Llama 3.2 is a large language model trained on the Llama 3.2 dataset.""",
tags=[],
metadata=ModelMetadata(
model_id="mlx-community/DeepSeek-R1-4bit",
pretty_name="DeepSeek R1 671B (4-bit)",
storage_size_kilobytes=409706307,
n_layers=61,
model_id="mlx-community/Llama-3.2-1B-Instruct-4bit",
pretty_name="Llama 3.2 1B",
storage_size_kilobytes=678948,
n_layers=16,
),
),
"deepseek-r1:671b": ModelCard(
short_id="deepseek-r1:671b",
model_id="mlx-community/DeepSeek-R1-4bit",
name="DeepSeek R1 671B",
description="""DeepSeek R1 is a large language model trained on the DeepSeek R1 dataset.""",
"llama-3.2-3b": ModelCard(
short_id="llama-3.2-3b",
model_id="mlx-community/Llama-3.2-3B-Instruct-4bit",
name="Llama 3.2 3B",
description="""Llama 3.2 is a large language model trained on the Llama 3.2 dataset.""",
tags=[],
metadata=ModelMetadata(
model_id="mlx-community/DeepSeek-R1-4bit",
pretty_name="DeepSeek R1 671B",
storage_size_kilobytes=409706307,
n_layers=61,
model_id="mlx-community/Llama-3.2-3B-Instruct-4bit",
pretty_name="Llama 3.2 3B",
storage_size_kilobytes=1765062,
n_layers=28,
),
),
"deepseek-v3": ModelCard(
short_id="deepseek-v3",
model_id="mlx-community/DeepSeek-V3-0324-4bit",
name="DeepSeek V3 4B",
description="""DeepSeek V3 is a large language model trained on the DeepSeek V3 dataset.""",
# llama-3.3
"llama-3.3-70b": ModelCard(
short_id="llama-3.3-70b",
model_id="mlx-community/Llama-3.3-70B-Instruct-4bit",
name="Llama 3.3 70B",
description="""The Meta Llama 3.3 multilingual large language model (LLM) is an instruction tuned generative model in 70B (text in/text out)""",
tags=[],
metadata=ModelMetadata(
model_id="mlx-community/DeepSeek-V3-0324-4bit",
pretty_name="DeepSeek V3 4B",
storage_size_kilobytes=368756663,
n_layers=61,
),
),
"deepseek-v3:671b": ModelCard(
short_id="deepseek-v3:671b",
model_id="mlx-community/DeepSeek-V3-0324-4bit",
name="DeepSeek V3 671B",
description="""DeepSeek V3 is a large language model trained on the DeepSeek V3 dataset.""",
tags=[],
metadata=ModelMetadata(
model_id="mlx-community/DeepSeek-V3-0324-4bit",
pretty_name="DeepSeek V3 671B",
storage_size_kilobytes=368756663,
n_layers=61,
model_id="mlx-community/Llama-3.3-70B-Instruct-4bit",
pretty_name="Llama 3.3 70B",
storage_size_kilobytes=38758160,
n_layers=80,
),
),
# phi-3
"phi-3-mini": ModelCard(
short_id="phi-3-mini",
model_id="mlx-community/Phi-3-mini-128k-instruct-4bit",
@@ -197,6 +170,8 @@ MODEL_CARDS: dict[str, ModelCard] = {
n_layers=32,
),
),
# qwen3
"qwen3-0.6b": ModelCard(
short_id="qwen3-0.6b",
model_id="mlx-community/Qwen3-0.6B-4bit",
@@ -223,6 +198,8 @@ MODEL_CARDS: dict[str, ModelCard] = {
n_layers=48,
),
),
# granite
"granite-3.3-2b": ModelCard(
short_id="granite-3.3-2b",
model_id="mlx-community/granite-3.3-2b-instruct-fp16",
@@ -249,6 +226,8 @@ MODEL_CARDS: dict[str, ModelCard] = {
n_layers=40,
),
),
# smol-lm
"smol-lm-135m": ModelCard(
short_id="smol-lm-135m",
model_id="mlx-community/SmolLM-135M-4bit",