SAVRN
Search Contact SAVRN

Open-weight model

kinyarwanda-mms-vits-finetune

by BANA Emmy Tresor tresorbana/kinyarwanda-mms-vits-finetune

This is the model card of a transformers model that has been pushed on the Hub. Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. Use the code below to get started with the model.

Parameters83M
Context
Weights5.3 GB
License
AccessOpen weights
Monthly Downloads50

Runs On

What it takes to serve kinyarwanda-mms-vits-finetune (83M parameters): the memory its weights need at each precision, and the cheapest way to rent enough data-center GPUs to hold them.

PrecisionWeightsMemory neededCheapest setupPer hourAlso fits
16-bit 0.2 GB 0.2 GB 1x MI300X (192 GB)
Vultr
$1.85 1x H100 $1.99 · 1x MI325X $2.00
8-bit 0.1 GB 0.1 GB 1x MI300X (192 GB)
Vultr
$1.85 1x H100 $1.99 · 1x MI325X $2.00
4-bit 0.0 GB 0.0 GB 1x MI300X (192 GB)
Vultr
$1.85 1x H100 $1.99 · 1x MI325X $2.00

Memory is the weights at that precision plus 20% for the runtime and a short context; a long context needs more. Prices are the lowest on-demand hourly rates in the SAVRN Index, read Sep 18, 2026.

Model Card

This is the model card of a transformers model that has been pushed on the Hub. Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. Use the code below to get started with the model. Carbon emissions can be estimated using the Machine Learning Impact calculator presented in Lacoste et al. (2019).

Excerpt from the card by BANA Emmy Tresor.

Configuration

Architecture
VitsModelForPreTraining
Layers
6
Hidden size
192
Attention heads
2
Vocabulary size
44
Stored precision
float32
Model type
vits

Identity and Version

Repository
tresorbana/kinyarwanda-mms-vits-finetune
Publisher
BANA Emmy Tresor
Task
Not stated by the source
Modality
Other
Library
transformers
Parameters
83M parameters
Languages
Not stated by the source
Revision
d479c453525cb1bfbb932ec84dea63cd4696b026
First published
2026-09-17
Last updated
2026-09-18

Files and Weights

54 files, 5.3 GB in total. The weights are 36 files totalling 5.3 GB in bin, pt, safetensors.

Weights36 files · 5.3 GB
Configuration4 files · 2.6 KB
Tokenizer2 files · 1.2 KB
Documentation1 file · 5.2 KB
Other10 files · 74.2 KB
Repository1 file · 1.5 KB
Every file
FileTypeSizeSHA-256
checkpoint-1500/model.safetensorsWeights145.3 MB 9d3183d1c6b2
checkpoint-1500/model_1.safetensorsWeights187.0 MB a06c6baa2224
checkpoint-1500/optimizer.binWeights291.1 MB 45ea549b2012
checkpoint-1500/optimizer_1.binWeights374.1 MB 45402e157520
checkpoint-1500/scaler.ptWeights1.4 KB edaca4d6d582
checkpoint-1500/scheduler.binWeights1.4 KB 14c933cedffa
checkpoint-1500/scheduler_1.binWeights1.4 KB 8cb397de7d7e
checkpoint-3000/model.safetensorsWeights145.3 MB e57d84b3cd37
checkpoint-3000/model_1.safetensorsWeights187.0 MB 2e470c4c461a
checkpoint-3000/optimizer.binWeights291.1 MB d65910be86a8
checkpoint-3000/optimizer_1.binWeights374.1 MB 926581dbcd1e
checkpoint-3000/scaler.ptWeights1.4 KB 4e785e7cd9a5
checkpoint-3000/scheduler.binWeights1.4 KB 896cf2caeae3
checkpoint-3000/scheduler_1.binWeights1.4 KB 74a516fd6834
checkpoint-4000/model.safetensorsWeights145.3 MB 866d19410278
checkpoint-4000/model_1.safetensorsWeights187.0 MB 48e27c298ae0
checkpoint-4000/optimizer.binWeights291.1 MB ae22edb93a05
checkpoint-4000/optimizer_1.binWeights374.1 MB 69d202cd2365
checkpoint-4000/scaler.ptWeights1.4 KB f2f9d30c1b8c
checkpoint-4000/scheduler.binWeights1.4 KB 3ffa670c9399
checkpoint-4000/scheduler_1.binWeights1.4 KB 225ef96674d2
checkpoint-6000/model.safetensorsWeights145.3 MB 27002fbb81b6
checkpoint-6000/model_1.safetensorsWeights187.0 MB 64990048a62b
checkpoint-6000/optimizer.binWeights291.1 MB e3c5ddf25a31
checkpoint-6000/optimizer_1.binWeights374.1 MB d81a22910479
checkpoint-6000/scaler.ptWeights1.4 KB f45a6c865320
checkpoint-6000/scheduler.binWeights1.4 KB 002514e7df60
checkpoint-6000/scheduler_1.binWeights1.4 KB 8f64e40260b3
checkpoint-8000/model.safetensorsWeights145.3 MB 4a3a2ef772b2
checkpoint-8000/model_1.safetensorsWeights187.0 MB e57287854a38
checkpoint-8000/optimizer.binWeights291.1 MB f19b16191930
checkpoint-8000/optimizer_1.binWeights374.1 MB 4d8560ac9625
checkpoint-8000/scaler.ptWeights1.4 KB 81c10b81995d
checkpoint-8000/scheduler.binWeights1.4 KB bd0ac7344ea2
checkpoint-8000/scheduler_1.binWeights1.4 KB 69491b36cf5d
model.safetensorsWeights332.2 MB 5846e8914e9d
added_tokens.jsonConfiguration18 B
config.jsonConfiguration2.1 KB
preprocessor_config.jsonConfiguration254 B
special_tokens_map.jsonConfiguration275 B
README.mdDocumentation5.2 KB
checkpoint-1500/random_states_0.pklOther14.8 KB 84fd7a739d76
checkpoint-3000/random_states_0.pklOther14.8 KB 6c2363bd4375
checkpoint-4000/random_states_0.pklOther14.8 KB 4b4ea5e7ce69
checkpoint-6000/random_states_0.pklOther14.9 KB a3fe45cf7d2b
checkpoint-8000/random_states_0.pklOther14.8 KB a68d1dd29700
step_1500.txtOther4 B
step_3000.txtOther4 B
step_4000.txtOther4 B
step_6000.txtOther4 B
step_8000.txtOther4 B
.gitattributesRepository1.5 KB
tokenizer_config.jsonTokenizer670 B
vocab.jsonTokenizer482 B

License and Download

License
Not stated by the source
Access
Open weights, no gate
Download size
5.3 GB
Download from BANA Emmy Tresor

Released by BANA Emmy Tresor through its official repository on Hugging Face.

Built From

Memory Requirements

PrecisionWeights in memory
As published5.3 GB
16-bit0.2 GB
8-bit0.1 GB
4-bit0.0 GB

Weights only, from the published parameter count; the key-value cache and runtime add to this.

Questions About kinyarwanda-mms-vits-finetune

How much GPU memory does kinyarwanda-mms-vits-finetune need?

About 0.2 GB at 16-bit and 0 GB at 4-bit: the weights (83M parameters) plus a working margin. A long context needs more.

What is the cheapest GPU to run kinyarwanda-mms-vits-finetune on?

At 16-bit, 1x MI300X from $1.85 an hour; at 4-bit, 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists.