Frontis-MA1-35B-SFT by Frontis AI: Open-Weight Model

Runs On

What it takes to serve Frontis-MA1-35B-SFT (36B parameters): the memory its weights need at each precision, and the cheapest way to rent enough data-center GPUs to hold them.

Precision	Weights	Memory needed	Cheapest setup	Per hour	Also fits
16-bit	71.9 GB	86.3 GB	1x MI300X (192 GB) Vultr	$1.85	1x MI325X $2.00 · 1x MI355X $2.59
8-bit	36.0 GB	43.1 GB	1x MI300X (192 GB) Vultr	$1.85	1x H100 $1.99 · 1x MI325X $2.00
4-bit	18.0 GB	21.6 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 Oct 19, 2026.

Frontis-MA1-35B-SFT on every accelerator the SAVRN Index prices, at every precision

Model Card

This is the supervised fine-tuning checkpoint of the Frontis-MA1 family, before OpenMLE reinforcement learning. It is intended for research on machine learning engineering and program transformation. The actual 24,350-row training input is distinct from the 26,259-example public SFT trace collection. Dataset cutoff configuration was 32,768 tokens; this does not establish validated long-context inference performance. No standalone SFT benchmark claim is made by this release. The separately released RL model is FrontisAI/Frontis-MA1-35B. Pin the published commit SHA for reproducibility. This model requires substantial memory; use the dependency versions recorded in the release validation…

Excerpt from the card by Frontis AI, licensed cc-by-nc-4.0.

Configuration

Architecture: Qwen3_5MoeForConditionalGeneration
Context length (tokens): 262,144
Layers: 40
Hidden size: 2,048
Attention heads: 16
Key/value heads: 2
Head dimension: 256
Vocabulary size: 248,320
Experts: 256
Experts active per token: 8
Model type: qwen3_5_moe

Identity and Version

Repository: FrontisAI/Frontis-MA1-35B-SFT
Publisher: Frontis AI
Task: Image and text to text
Modality: Image and text
Library: transformers
Parameters: 36B parameters
Languages: en
Revision: 2c622878bea76bfcd130a879fbe9618708bd35f8
First published: 2026-10-15
Last updated: 2026-10-16

Files and Weights

33 files, 71.9 GB in total. The weights are 15 files totalling 71.9 GB in safetensors.

Weights15 files · 71.9 GB

Configuration7 files · 107.5 KB

Tokenizer4 files · 22.9 MB

Documentation4 files · 16.6 KB

Other2 files · 10.6 KB

Repository1 file · 1.6 KB

Every file

File	Type	Size	SHA-256
model-00000-of-00015.safetensors	Weights	5.3 GB	ce7be386c63d
model-00001-of-00015.safetensors	Weights	4.9 GB	45fd7d49410f
model-00002-of-00015.safetensors	Weights	4.9 GB	93e08e89022f
model-00003-of-00015.safetensors	Weights	4.9 GB	5018a305b746
model-00004-of-00015.safetensors	Weights	4.9 GB	0134d1135a80
model-00005-of-00015.safetensors	Weights	4.9 GB	8e7836bb3dde
model-00006-of-00015.safetensors	Weights	4.9 GB	dc9780b253c2
model-00007-of-00015.safetensors	Weights	4.9 GB	5437e172f3a5
model-00008-of-00015.safetensors	Weights	4.9 GB	cdb7d65981d3
model-00009-of-00015.safetensors	Weights	4.9 GB	b8632cd71fb2
model-00010-of-00015.safetensors	Weights	4.9 GB	a33e510c2894
model-00011-of-00015.safetensors	Weights	4.9 GB	549e8e2397b2
model-00012-of-00015.safetensors	Weights	4.9 GB	89c28f7b202f
model-00013-of-00015.safetensors	Weights	5.4 GB	9334e9249bae
model-00014-of-00015.safetensors	Weights	2.7 GB	f39e759c4056
config.json	Configuration	3.7 KB	—
configuration.json	Configuration	58 B	—
generation_config.json	Configuration	202 B	—
model.safetensors.index.json	Configuration	98.4 KB	—
preprocessor_config.json	Configuration	390 B	—
release_manifest.json	Configuration	4.4 KB	—
video_preprocessor_config.json	Configuration	385 B	—
LICENSE	Documentation	1.1 KB	—
LICENSE-UPSTREAM-APACHE-2.0	Documentation	11.3 KB	—
NOTICE	Documentation	1.2 KB	—
README.md	Documentation	3.0 KB	—
chat_template.jinja	Other	7.8 KB	—
checksums.sha256	Other	2.8 KB	—
.gitattributes	Repository	1.6 KB	—
merges.txt	Tokenizer	3.4 MB	—
tokenizer.json	Tokenizer	12.8 MB	5f9e4d4901a9
tokenizer_config.json	Tokenizer	16.7 KB	—
vocab.json	Tokenizer	6.7 MB	—

License and Download

License: cc-by-nc-4.0
Access: Open weights, no gate
Download size: 71.9 GB

Download from Frontis AI

Released by Frontis AI through its official repository on Hugging Face. Read the license.

Built From

Derived from Qwen/Qwen3.6-35B-A3B

Memory Requirements

Precision	Weights in memory
As published	71.9 GB
16-bit	71.9 GB
8-bit	36.0 GB
4-bit	18.0 GB

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

Questions About Frontis-MA1-35B-SFT

How much GPU memory does Frontis-MA1-35B-SFT need?

About 86.3 GB at 16-bit and 21.6 GB at 4-bit: the weights (36B parameters) plus a working margin. A long context needs more.

What is the cheapest GPU to run Frontis-MA1-35B-SFT 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.

Can I use Frontis-MA1-35B-SFT commercially?

Not without separate permission. Frontis-MA1-35B-SFT is released under Creative Commons Attribution-NonCommercial 4.0. CC BY-NC 4.0 permits sharing and adapting with credit for non-commercial purposes only. Commercial use needs separate permission from the rights holder.

What is Frontis-MA1-35B-SFT's context length?

262,144 tokens, from the maximum position embeddings in its published configuration.

Similar Models

Model · Image and text to text

Qwen3.6-35B-A3B

Qwen

Following the February release of the Qwen3.5 series, we're pleased to share the first open-weight variant of Qwen3.6. Built on direct feedback from the community, Qwen3.6 prioritizes stability and real-world utility, offering developers a more intuitive, responsive, and genuinely productive coding experience. This release delivers substantial upgrades, particularly in For more details, please refer to our blog post Qwen3.6-35B-A3B. Empty cells (--) indicate scores not available or not applicable. For streamlined integration, we recommend using Qwen3.6 via APIs. Below is a guide to use Qwen3.6 via OpenAI-compatible API. Qwen3.6 can be served via APIs with popular inference frameworks. In…

Open weights apache-2.0 36B parameters 262,144 tokens transformers

View model

Model · Image and text to text

Qwen3.5-35B-A3B

Qwen

Over recent months, we have intensified our focus on developing foundation models that deliver exceptional utility and performance. Qwen3.5 represents a significant leap forward, integrating breakthroughs in multimodal learning, architectural efficiency, reinforcement learning scale, and global accessibility to empower developers and enterprises with unprecedented capability and efficiency. For more details, please refer to our blog post Qwen3.5. WMT24++: a harder subset of WMT24 after difficulty labeling and rebalancing; we report the averaged scores on 55 languages using XCOMET-XXL. Empty cells (--) indicate scores not yet available or not applicable. Empty cells (--) indicate scores not…

Open weights apache-2.0 36B parameters 262,144 tokens transformers

View model

Model · Image and text to text

Qwen3.6-35B-A3B-Brainwaves-Nex-Uncensored

Gheorghe Chesler

This model is a merge of: - nightmedia/Qwen3.6-35B-A3B-Brainwaves - orcarouter/Nex-N2.5-mini-Uncensored - thomsonreuters/Thomson-1.0-Small - nightmedia/Qwen3.6-35B-A3B-Fable-Holo3.1 - nightmedia/Qwen3.6-35B-A3B-FSM - orcarouter/Nex-N2.5-mini-Uncensored Lab name: Qwen3.6-35B-A3B-AgentWorld-Fable-Holo3.1-Iris-Thomson-B-Nex2 Brainwaves Performance with different expert count (more metrics coming) Qwen3.6-35B-A3B-Brainwaves orcarouter/Nex-N2.5-mini-Uncensored The merge with the original nex-agi/Nex-N2.5-mini Qwen3.6-35B-A3B-Brainwaves-Nex nex-agi/Nex-N2.5-mini Qwen3.6-35B-A3B-Instruct This model is using(an early version of) the fixed jinja template from froggeric/Qwen-Fixed-Chat-Templates Drop…

Open weights apache-2.0 36B parameters 262,144 tokens transformers

View model

Model · Image and text to text

qwen3.6-35b-a3b-tool-prompts-ckpt444-merged-fp8

Manish

Base: Qwen/Qwen3.6-35B-A3B-FP8 (same FP8 block format/scales convention). Adapter: qwen3.6-35b-a3b-tool-prompts-ckpt444-adapter (LoRA r=32, 2 epochs, 1873 dialogues). Merged in BF16 (RAM only), block-requantized to FP8 e4m3/128. Chat template: qwen3template333.jinja. Serve with sglang / vllm / transformers.

Open weights apache-2.0 36B parameters 262,144 tokens

View model

Model · Image and text to text

qwen3p6-35b-a3b-pin-2p02bit-r46

MinJae Choi

Following the February release of the Qwen3.5 series, we're pleased to share the first open-weight variant of Qwen3.6. Built on direct feedback from the community, Qwen3.6 prioritizes stability and real-world utility, offering developers a more intuitive, responsive, and genuinely productive coding experience. This release delivers substantial upgrades, particularly in For more details, please refer to our blog post Qwen3.6-35B-A3B. Empty cells (--) indicate scores not available or not applicable. For streamlined integration, we recommend using Qwen3.6 via APIs. Below is a guide to use Qwen3.6 via OpenAI-compatible API. Qwen3.6 can be served via APIs with popular inference frameworks. In…

Open weights apache-2.0 36B parameters 262,144 tokens transformers

View model

Model · Image and text to text

Qwen3.6-35B-A3B-oQ6e-mtp

Chris Derby

oQ6e mixed-precision MLX quantization of Qwen/Qwen3.6-35B-A3B, made with oMLX oQe (v0.7.1.dev1). This is the bf16 build. The multi-token-prediction (MTP) head is kept and the vision tower is included. The same family at every bit width and dtype (this repo in bold): Which dtype? bf16 (no suffix) is the default. fp16 builds store the language model's non-quantized weights and the quantization scales as float16 instead of bfloat16. They are marginally closer to the original at every level. oMLX notes that float16 can give faster prefill on M1/M2, which lack native bfloat16. They are about 0.9 GB larger because the vision tower is kept in float32; vision quality was not measured. How these…

Open weights apache-2.0 36B parameters 262,144 tokens mlx

View model