3b-de-ft-research_release-4bit

mlx-community/3b-de-ft-research_release-4bit

Text to speech · transformers

4-bit Open on Hugging Face

Parameters

3.3B

Weights (est.)

≈ 1.9 GB

Mac memory (est.)

8 GB+

Size is estimated from the parameter count and quantisation. Memory allows a fifth extra for context plus 2 GB for the system; long contexts need more.

Details

Base model: canopylabs/orpheus-3b-0.1-pretrained (finetune)
Vendor: canopylabs
Task: Text to speech
Quantisation: 4-bit
Licence: apache-2.0
Downloads: 82 in the last 30 days · 938 all time · 0 likes
Published: 2025-04-09 · updated 2025-04-09 (1y ago)
Languages: de
Tags: llama, text-generation

Other MLX builds of canopylabs/orpheus-3b-0.1-pretrained

Build	Quantisation	Weights	Mac memory	Downloads
orpheus-3b-0.1-pretrained-4bit	4-bit	≈ 1.9 GB	8 GB+	30
3b-de-ft-research_release-4bit (this one)	4-bit	≈ 1.9 GB	8 GB+	82
3b-es_it-ft-research_release-4bit	4-bit	≈ 1.9 GB	8 GB+	74
3b-fr-ft-research_release-4bit	4-bit	≈ 1.9 GB	8 GB+	49
3b-zh-ft-research_release-4bit	4-bit	≈ 1.9 GB	8 GB+	40
3b-ko-ft-research_release-4bit	4-bit	≈ 1.9 GB	8 GB+	39
3b-hi-ft-research_release-4bit	4-bit	≈ 1.9 GB	8 GB+	28
3b-hi-pretrain-research_release-4bit	4-bit	≈ 1.9 GB	8 GB+	22
orpheus-3b-0.1-pretrained-6bit	6-bit	≈ 2.7 GB	8 GB+	22
3b-de-ft-research_release-6bit	6-bit	≈ 2.7 GB	8 GB+	98
3b-zh-ft-research_release-6bit	6-bit	≈ 2.7 GB	8 GB+	69
3b-fr-ft-research_release-6bit	6-bit	≈ 2.7 GB	8 GB+	46
3b-es_it-ft-research_release-6bit	6-bit	≈ 2.7 GB	8 GB+	37
3b-ko-ft-research_release-6bit	6-bit	≈ 2.7 GB	8 GB+	31
3b-hi-ft-research_release-6bit	6-bit	≈ 2.7 GB	8 GB+	29
3b-hi-pretrain-research_release-6bit	6-bit	≈ 2.7 GB	8 GB+	13
orpheus-3b-0.1-pretrained-8bit	8-bit	≈ 3.5 GB	8 GB+	33
3b-de-ft-research_release-8bit	8-bit	≈ 3.5 GB	8 GB+	94
3b-fr-ft-research_release-8bit	8-bit	≈ 3.5 GB	8 GB+	79
3b-es_it-ft-research_release-8bit	8-bit	≈ 3.5 GB	8 GB+	71
3b-ko-ft-research_release-8bit	8-bit	≈ 3.5 GB	8 GB+	68
3b-hi-ft-research_release-8bit	8-bit	≈ 3.5 GB	8 GB+	53
3b-zh-ft-research_release-8bit	8-bit	≈ 3.5 GB	8 GB+	35
3b-hi-pretrain-research_release-8bit	8-bit	≈ 3.5 GB	8 GB+	18
orpheus-3b-0.1-pretrained-bf16	bf16	≈ 6.6 GB	16 GB+	20
3b-de-ft-research_release-bf16	bf16	—	—	87
3b-es_it-ft-research_release-bf16	bf16	—	—	70
3b-fr-ft-research_release-bf16	bf16	—	—	68
3b-zh-ft-research_release-bf16	bf16	—	—	64
3b-ko-ft-research_release-bf16	bf16	—	—	53
3b-hi-ft-research_release-bf16	bf16	—	—	25
3b-hi-pretrain-research_release-bf16	bf16	—	—	12