# matilda-jev-v1 by Maincode: Open-Weight Model Source: https://savrn.com/models/matilda-jev-v1 Markdown alternate of the page above; the site index is https://savrn.com/llms.txt --- ## Runs On What it takes to serve matilda-jev-v1 (26.1B 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 | 52.2 GB | 62.6 GB | 1x [MI300X](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/mi325x) $2.00 | | 8-bit | 26.1 GB | 31.3 GB | 1x [MI300X](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/mi325x) $2.00 | | 4-bit | 13.0 GB | 15.7 GB | 1x [MI300X](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/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](https://savrn.com/ai-index/pricing/gpus), read Oct 7, 2026. [matilda-jev-v1 on every accelerator the SAVRN Index prices, at every precision](https://savrn.com/models/matilda-jev-v1/gpus) ## Model Card By Maincode, published under apache-2.0, revision 3dd1a9db4c33. ### Matilda-Jev by Maincode Matilda-Jev is Maincode's one-pass decision model. It scores the options supplied in a choice, noul (yes/no), or ordered score question. It accepts text or JSON state and optional images. This is a decision checkpoint with a 255-option readout, not a text-generation checkpoint. This configuration edition uses MatildaJevModel, MatildaJevConfig, and MATILDA tokenizer/processor classes. Use the bundled runtime below, or load the custom AutoClasses with trust_remote_code=True. The package includes the backbone weights, decision readout, tokenizer, preprocessing configuration and calibrated temperature. ### Validation On AMD MI355X with Python 3.12, Transformers 5.17.0 and PyTorch 2.14.0: - 25/25 smoke-test questions passed, including six product-identity questions and an image question. - All 23 requests matched the original checkpoint's answer probabilities exactly (maximum difference 0). - Tokenization, image preprocessing, configuration save/reload and architecture-parameter comparisons passed. - A real forward/backward pass produced finite, nonzero readout and embedding gradients. No optimizer update was applied. [Read the full model card (474 words)](https://savrn.com/models/matilda-jev-v1/card) ## Configuration Architecture MatildaJevModel Context length (tokens) 262,144 Layers 64 Hidden size 5,120 Feed-forward size 17,408 Attention heads 24 Key/value heads 4 Head dimension 256 Vocabulary size 248,320 Model type matilda_jev ## Identity and Version Repository Maincode/matilda-jev-v1 Publisher Maincode Task Feature extraction Modality Text Library transformers Parameters 26.1B parameters Languages en Revision 3dd1a9db4c337f6aca80d4b682a1fcf712944d7e First published 2026-09-30 Last updated 2026-10-07 ## Files and Weights 41 files, 52.2 GB in total. The weights are 12 files totalling 52.2 GB in safetensors. Weights12 files · 52.2 GB Configuration20 files · 171.9 KB Tokenizer2 files · 20.0 MB Documentation2 files · 16.1 KB Other4 files · 33.1 KB Repository1 file · 1.6 KB Every file | File | Type | Size | SHA-256 | | --- | --- | --- | --- | | model-00001-of-00011.safetensors | Weights | 5.0 GB | ad9615133ae2 | | model-00002-of-00011.safetensors | Weights | 5.0 GB | f1488d845eec | | model-00003-of-00011.safetensors | Weights | 4.9 GB | 53d90ef212d0 | | model-00004-of-00011.safetensors | Weights | 5.0 GB | 2288a0f3d9dd | | model-00005-of-00011.safetensors | Weights | 5.0 GB | e7a08e2b452a | | model-00006-of-00011.safetensors | Weights | 4.9 GB | 688723f918f8 | | model-00007-of-00011.safetensors | Weights | 5.0 GB | 9251e36e82d2 | | model-00008-of-00011.safetensors | Weights | 4.9 GB | 59bf8ac0ae5b | | model-00009-of-00011.safetensors | Weights | 5.0 GB | cc010e178005 | | model-00010-of-00011.safetensors | Weights | 4.9 GB | 95b271104f64 | | model-00011-of-00011.safetensors | Weights | 2.6 GB | 8b17d96a9816 | | readout.safetensors | Weights | 2.6 MB | 262b432eea8c | | CPU_TEST.json | Configuration | 370 B | — | | PACKAGE_MANIFEST.json | Configuration | 6.0 KB | — | | TEST_REPORT.json | Configuration | 2.5 KB | — | | WEIGHTS_VERIFIED.json | Configuration | 2.4 KB | — | | config.json | Configuration | 3.9 KB | — | | configuration_matilda_jev.py | Configuration | 603 B | — | | decision_config.json | Configuration | 5.4 KB | — | | model.safetensors.index.json | Configuration | 104.0 KB | — | | modeling_matilda_jev.py | Configuration | 1.3 KB | — | | processing_matilda_jev.py | Configuration | 1.5 KB | — | | processor_config.json | Configuration | 1.3 KB | — | | runtime/maincode_jev_serve/__init__.py | Configuration | | — | | runtime/maincode_jev_serve/ask.py | Configuration | 4.8 KB | — | | runtime/maincode_jev_serve/config.py | Configuration | 2.1 KB | — | | runtime/maincode_jev_serve/decide.py | Configuration | 3.7 KB | — | | runtime/maincode_jev_serve/engine.py | Configuration | 3.3 KB | — | | runtime/maincode_jev_serve/model.py | Configuration | 14.4 KB | — | | runtime/maincode_jev_serve/server.py | Configuration | 12.2 KB | — | | runtime/maincode_jev_serve/types.py | Configuration | 1.9 KB | — | | tokenization_matilda_jev.py | Configuration | 280 B | — | | LICENSE | Documentation | 11.5 KB | — | | README.md | Documentation | 4.5 KB | — | | USAGE.txt | Other | 2.4 KB | — | | chat_template.jinja | Other | 9.0 KB | — | | requirements-runtime.txt | Other | 220 B | — | | runtime/maincode_jev_serve/playground.html | Other | 21.5 KB | — | | .gitattributes | Repository | 1.6 KB | — | | tokenizer.json | Tokenizer | 20.0 MB | 6f32ce20dc35 | | tokenizer_config.json | Tokenizer | 1.5 KB | — | ## License and Download License apache-2.0 Access Open weights, no gate Download size 52.2 GB [Download from Maincode](https://huggingface.co/Maincode/matilda-jev-v1) Released by Maincode through its official repository on Hugging Face. [Read the license](https://www.apache.org/licenses/LICENSE-2.0). ## Built From - Derived from [Qwen/Qwen3.8-27B](https://savrn.com/models/qwen3-8-27b) ## Memory Requirements | Precision | Weights in memory | | --- | --- | | As published | 52.2 GB | | 16-bit | 52.2 GB | | 8-bit | 26.1 GB | | 4-bit | 13.0 GB | Weights only, from the published parameter count; the key-value cache and runtime add to this. ## Questions About matilda-jev-v1 ### How much GPU memory does matilda-jev-v1 need? About 62.6 GB at 16-bit and 15.7 GB at 4-bit: the weights (26.1B parameters) plus a working margin. A long context needs more. ### What is the cheapest GPU to run matilda-jev-v1 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 matilda-jev-v1 commercially? Yes. matilda-jev-v1 is released under Apache License 2.0. The Apache License 2.0 is a permissive open-source license. It permits commercial use, modification and redistribution. It requires keeping the license and copyright notices and any NOTICE file, stating significant changes, and it includes an express patent grant from contributors. ### What is matilda-jev-v1's context length? 262,144 tokens, from the maximum position embeddings in its published configuration. ## Similar Models Model · Feature extraction ### [webshop_8b_b8_1node_step150](https://savrn.com/models/webshop-8b-b8-1node-step150) [Longyy](https://savrn.com/model-publishers/loongyy) This model is a fine-tuned version of JetLM/SDAR-8B-Chat-b8 on the sdarwebshopbs4eighthreason dataset. The following hyperparameters were used during training: - learningrate: 1e-05 - trainbatchsize: 1 - evalbatchsize: 8 - distributedtype: multi-GPU - numdevices: 4 - totaltrainbatchsize: 4 - totalevalbatchsize: 32 - lrschedulertype: constantwithwarmup - lrschedulerwarmupratio: 0.03 - numepochs: 1.0 - Transformers 4.52.4 - Pytorch 2.9.1+cu129 - Datasets 3.6.0 - Tokenizers 0.21.1 Open weights other 8.2B parameters 32,768 tokens transformers [View model](https://savrn.com/models/webshop-8b-b8-1node-step150) Model · Feature extraction ### [CLM-v0.1-8B-MLX-8bit](https://savrn.com/models/clm-v0-1-8b-mlx-8bit) [Zhu Lin](https://savrn.com/model-publishers/czl) This is the 8bit variant. Also available: czl/CLM-v0.1-8B-MLX-4bit, czl/CLM-v0.1-8B-MLX-6bit, and the bf16 reference. The encoder half of Contrastive-LM/CLM-v0.1-8B, quantised for MLX on Apple Silicon. mlxlm is generate-only — it has no embeddings entrypoint, and mlxlm.server routes just /v1/completions, /v1/chat/completions, /v1/models and /health. So clmmlx supplies the pooling pass over mlxlm internals: Qwen3Model.call already returns self.norm(h), the post-final-RMSNorm hidden states, which is what vLLM's pooling runner returns in last-token mode. Agreement against a bf16 MLX reference of the same encoder in the same runtime, over 23,926 scored System One questions, through the real… Open weights apache-2.0 8.2B parameters 40,960 tokens mlx [View model](https://savrn.com/models/clm-v0-1-8b-mlx-8bit) Model · Feature extraction ### [CLM-v0.1-8B-MLX](https://savrn.com/models/clm-v0-1-8b-mlx) [Zhu Lin](https://savrn.com/model-publishers/czl) The encoder half of Contrastive-LM/CLM-v0.1-8B, quantised for MLX on Apple Silicon. One repository per bit width, matching mlx-community. Each has the weights at the repo root, so the Hub file browser lists every file with its size and mlxlm.load(" ") works. Agreement against a bf16 MLX reference of the same encoder in the same runtime, over 23,926 scored System One questions, through the real head stack (argmax(scale · cos), scale = 100.0 — cosine error is amplified 100×). Pass line. top-1 >= 1.0000 — the measured bf16-vs-bf16 noise floor of this corpus in this runtime — and top-1 (decisive) >= 0.995, where decisive means the reference's own top-1 led by more than 1 nat. A third condition… Open weights apache-2.0 8.2B parameters 40,960 tokens mlx [View model](https://savrn.com/models/clm-v0-1-8b-mlx) Model · Feature extraction ### [CLM-v0.1-8B-MLX-6bit](https://savrn.com/models/clm-v0-1-8b-mlx-6bit) [Zhu Lin](https://savrn.com/model-publishers/czl) This is the 6bit variant. Also available: czl/CLM-v0.1-8B-MLX-4bit, czl/CLM-v0.1-8B-MLX-8bit, and the bf16 reference. The encoder half of Contrastive-LM/CLM-v0.1-8B, quantised for MLX on Apple Silicon. mlxlm is generate-only — it has no embeddings entrypoint, and mlxlm.server routes just /v1/completions, /v1/chat/completions, /v1/models and /health. So clmmlx supplies the pooling pass over mlxlm internals: Qwen3Model.call already returns self.norm(h), the post-final-RMSNorm hidden states, which is what vLLM's pooling runner returns in last-token mode. Agreement against a bf16 MLX reference of the same encoder in the same runtime, over 23,926 scored System One questions, through the real… Open weights apache-2.0 8.2B parameters 40,960 tokens mlx [View model](https://savrn.com/models/clm-v0-1-8b-mlx-6bit) Model · Feature extraction ### [Qwen3-Embedding-8B](https://savrn.com/models/qwen3-embedding-8b) [Qwen](https://savrn.com/model-publishers/qwen) The Qwen3 Embedding model series is the latest proprietary model of the Qwen family, specifically designed for text embedding and ranking tasks. Building upon the dense foundational models of the Qwen3 series, it provides a comprehensive range of text embeddings and reranking models in various sizes (0.6B, 4B, and 8B). This series inherits the exceptional multilingual capabilities, long-text understanding, and reasoning skills of its foundational model. The Qwen3 Embedding series represents significant advancements in multiple text embedding and ranking tasks, including text retrieval, code retrieval, text classification, text clustering, and bitext mining. Exceptional Versatility: The… Open weights apache-2.0 7.6B parameters 40,960 tokens sentence-transformers [View model](https://savrn.com/models/qwen3-embedding-8b) Model · Feature extraction ### [e5-mistral-7b-instruct-bnb-4bit](https://savrn.com/models/e5-mistral-7b-instruct-bnb-4bit) [Gábor Hosu](https://savrn.com/model-publishers/gabor-hosu) This model is a quantized version of the original model intfloat/e5-mistral-7b-instruct. It's quantized using the BitsAndBytes library to 4-bit using the bnb-my-repo space. - bnb4bitquanttype: nf4 - bnb4bitusedoublequant: True - bnb4bitcomputedtype: bfloat16 - bnb4bitquantstorage: uint8 Improving Text Embeddings with Large Language Models. Liang Wang, Nan Yang, Xiaolong Huang, Linjun Yang, Rangan Majumder, Furu Wei, arXiv 2024 This model has 32 layers and the embedding size is 4096. Below is an example to encode queries and passages from the MS-MARCO passage ranking dataset. Have a look at configsentencetransformers.json for the prompts that are pre-configured, such as websearchquery… Open weights mit 7.3B parameters 32,768 tokens sentence-transformers [View model](https://savrn.com/models/e5-mistral-7b-instruct-bnb-4bit) ## Maincode [All models and datasets](https://savrn.com/model-publishers/maincode) ## Versions - [3dd1a9db4c33](https://savrn.com/models/matilda-jev-v1/versions/3dd1a9db4c33) · current 2026-10-07 ## Explore More - [All feature extraction models](https://savrn.com/models/tasks/feature-extraction) - [All models under apache-2.0](https://savrn.com/models/licenses/apache-2-0) - [Model comparisons](https://savrn.com/models/comparisons) - [The model directory](https://savrn.com/models) - [Open model prices by host](https://savrn.com/ai-index/pricing/open-models) ## Source - Repository metadata, read 2026-10-07. - [Hugging Face record](https://huggingface.co/Maincode/matilda-jev-v1) - [How the hub is built](https://savrn.com/model-hub/methodology)