MLX 6-bit weights with group size 64; a 3.732 GB weight package. Fixed memory for long-running generation. This EdgeRock-4B deployment package retains recurrent state, local attention, and Rock Memory Anchors (RMA), keeping persistent history at a fixed capacity as context grows. 固定记忆，长程生成。 本版本保留 EdgeRock-4B 的递推状态、局部注意力和记忆岩锚（RMA），历史内存不随上下文增长。 Rock Matrix Whitepaper / 岩芯矩阵技术白皮书： whitepaper connects EdgeRock's fixed memory, DiffusionRock's parallel generation, and SiliconRock's model-native silicon into the route toward Ultra-Fast Token. 白皮书讲清固定记忆、并行生成、模型固化三条路线，以及它们如何走向更快、更经济的生成。 The instructions and validation below apply to this package. Browse all EdgeRock models. Original-model benchmarks…
Open-weight model · Text generation
EdgeRock-4B-MLX-8bit
by Aaron Xu hkust-aaron/EdgeRock-4B-MLX-8bit
EdgeRock-4B-MLX-8bit is an open-weight model for text generation from Aaron Xu, released under other. It has 4.6B parameters and a 262,144-token context. At 16-bit it needs about 11 GB of GPU memory, which fits on 1x MI300X from $1.85 an hour, at the lowest prices in the SAVRN Index. It draws 60 downloads a month.
MLX 8-bit weights with group size 64; a 4.874 GB weight package. Fixed memory for long-running generation.
Parameters4.6B
Context262,144
Weights4.9 GB
Licenseother
AccessOpen weights
Monthly Downloads60
Runs On
What it takes to serve EdgeRock-4B-MLX-8bit (4.6B 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 | 9.2 GB | 11.0 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 8-bit | 4.6 GB | 5.5 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 4-bit | 2.3 GB | 2.7 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 23, 2026.
EdgeRock-4B-MLX-8bit on every accelerator the SAVRN Index prices, at every precision
Model Card
MLX 8-bit weights with group size 64; a 4.874 GB weight package. Fixed memory for long-running generation. This EdgeRock-4B deployment package retains recurrent state, local attention, and Rock Memory Anchors (RMA), keeping persistent history at a fixed capacity as context grows. 固定记忆，长程生成。 本版本保留 EdgeRock-4B 的递推状态、局部注意力和记忆岩锚（RMA），历史内存不随上下文增长。 Rock Matrix Whitepaper / 岩芯矩阵技术白皮书： whitepaper connects EdgeRock's fixed memory, DiffusionRock's parallel generation, and SiliconRock's model-native silicon into the route toward Ultra-Fast Token. 白皮书讲清固定记忆、并行生成、模型固化三条路线，以及它们如何走向更快、更经济的生成。 The instructions and validation below apply to this package. Browse all EdgeRock models. Original-model benchmarks…
Excerpt from the card by Aaron Xu, licensed other.
Configuration
- Architecture
- EdgeRockForCausalLM
- Context length (tokens)
- 262,144
- Layers
- 32
- Hidden size
- 2,560
- Feed-forward size
- 9,216
- Attention heads
- 16
- Key/value heads
- 4
- Head dimension
- 256
- Vocabulary size
- 248,320
- Model type
- edgerock
Identity and Version
- Repository
- hkust-aaron/EdgeRock-4B-MLX-8bit
- Publisher
- Aaron Xu
- Task
- Text generation
- Modality
- Text
- Library
- mlx
- Parameters
- 4.6B parameters
- Languages
- mlx
- Revision
- 5b6e379b16c596f7bcb1c8defa6151ada91bba04
- First published
- 2026-10-17
- Last updated
- 2026-10-23
Files and Weights
15 files, 4.9 GB in total. The weights are 1 file totalling 4.9 GB in safetensors.
Weights1 file · 4.9 GB
Configuration6 files · 145.3 KB
Tokenizer2 files · 20.0 MB
Documentation4 files · 71.4 KB
Other1 file · 115 B
Repository1 file · 1.6 KB
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| model.safetensors | Weights | 4.9 GB | bafcca551879 |
| WEIGHTS.json | Configuration | 243 B | — |
| config.json | Configuration | 2.9 KB | — |
| edgerock_mlx.py | Configuration | 8.6 KB | — |
| generation_config.json | Configuration | 120 B | — |
| receipts/MLX-8bit.json | Configuration | 132.3 KB | — |
| run_mlx.py | Configuration | 1.2 KB | — |
| LICENSE | Documentation | 18.0 KB | — |
| LICENSE-CODE | Documentation | 11.4 KB | — |
| NOTICE | Documentation | 3.6 KB | — |
| README.md | Documentation | 38.4 KB | — |
| requirements.txt | Other | 115 B | — |
| .gitattributes | Repository | 1.6 KB | — |
| tokenizer.json | Tokenizer | 20.0 MB | 06b9509352d2 |
| tokenizer_config.json | Tokenizer | 1.1 KB | — |
License and Download
- License
- other
- Access
- Open weights, no gate
- Download size
- 4.9 GB
Download from Aaron Xu
Released by Aaron Xu through its official repository on Hugging Face.
Built From
- Derived from rockmatrix-ai/EdgeRock
- Described by arXiv:2610.10118
- Quantized from rockmatrix-ai/EdgeRock
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 4.9 GB |
| 16-bit | 9.2 GB |
| 8-bit | 4.6 GB |
| 4-bit | 2.3 GB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.
Questions About EdgeRock-4B-MLX-8bit
How much GPU memory does EdgeRock-4B-MLX-8bit need?
About 11 GB at 16-bit and 2.7 GB at 4-bit: the weights (4.6B parameters) plus a working margin. A long context needs more.
What is the cheapest GPU to run EdgeRock-4B-MLX-8bit 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.
What license is EdgeRock-4B-MLX-8bit released under?
other, as its publisher declares it. Read the license text before commercial use.
What is EdgeRock-4B-MLX-8bit's context length?
262,144 tokens, from the maximum position embeddings in its published configuration.
Similar Models
MLX 4-bit weights with group size 32; a 2.875 GB weight package. Fixed memory for long-running generation. This EdgeRock-4B deployment package retains recurrent state, local attention, and Rock Memory Anchors (RMA), keeping persistent history at a fixed capacity as context grows. 固定记忆，长程生成。 本版本保留 EdgeRock-4B 的递推状态、局部注意力和记忆岩锚（RMA），历史内存不随上下文增长。 Rock Matrix Whitepaper / 岩芯矩阵技术白皮书： whitepaper connects EdgeRock's fixed memory, DiffusionRock's parallel generation, and SiliconRock's model-native silicon into the route toward Ultra-Fast Token. 白皮书讲清固定记忆、并行生成、模型固化三条路线，以及它们如何走向更快、更经济的生成。 The instructions and validation below apply to this package. Browse all EdgeRock models. Original-model benchmarks…
MLX mixed precision: 4-bit MLPs and 8-bit other eligible matrices, with group size 64; 3.741 GB of weights. Fixed memory for long-running generation. This EdgeRock-4B deployment package retains recurrent state, local attention, and Rock Memory Anchors (RMA), keeping persistent history at a fixed capacity as context grows. 固定记忆，长程生成。 本版本保留 EdgeRock-4B 的递推状态、局部注意力和记忆岩锚（RMA），历史内存不随上下文增长。 Rock Matrix Whitepaper / 岩芯矩阵技术白皮书： whitepaper connects EdgeRock's fixed memory, DiffusionRock's parallel generation, and SiliconRock's model-native silicon into the route toward Ultra-Fast Token. 白皮书讲清固定记忆、并行生成、模型固化三条路线，以及它们如何走向更快、更经济的生成。 The instructions and validation below apply to this package. Browse all…
AWQ W4A16 weights for Rock Matrix's EdgeRock-4B, with a packed INT4 CUDA runtime. Fixed memory for long-running generation. This EdgeRock-4B deployment package retains recurrent state, local attention, and Rock Memory Anchors (RMA), keeping persistent history at a fixed capacity as context grows. 固定记忆，长程生成。 本版本保留 EdgeRock-4B 的递推状态、局部注意力和记忆岩锚（RMA），历史内存不随上下文增长。 Rock Matrix Whitepaper / 岩芯矩阵技术白皮书： whitepaper connects EdgeRock's fixed memory, DiffusionRock's parallel generation, and SiliconRock's model-native silicon into the route toward Ultra-Fast Token. 白皮书讲清固定记忆、并行生成、模型固化三条路线，以及它们如何走向更快、更经济的生成。 EdgeRock combines recurrent state, sliding-window attention and Rock Memory Anchors (RMA) to…
GPTQ W4A16 weights for Rock Matrix's EdgeRock-4B, with a packed INT4 CUDA runtime. Fixed memory for long-running generation. This EdgeRock-4B deployment package retains recurrent state, local attention, and Rock Memory Anchors (RMA), keeping persistent history at a fixed capacity as context grows. 固定记忆，长程生成。 本版本保留 EdgeRock-4B 的递推状态、局部注意力和记忆岩锚（RMA），历史内存不随上下文增长。 Rock Matrix Whitepaper / 岩芯矩阵技术白皮书： whitepaper connects EdgeRock's fixed memory, DiffusionRock's parallel generation, and SiliconRock's model-native silicon into the route toward Ultra-Fast Token. 白皮书讲清固定记忆、并行生成、模型固化三条路线，以及它们如何走向更快、更经济的生成。 EdgeRock combines recurrent state, sliding-window attention and Rock Memory Anchors (RMA) to…
YANchor-4B is a general-purpose recurrent model that preserves crucial memory as ANchors for retrieval during subsequent reasoning. Its multidimensional memory mechanism enables effective long-horizon reasoning with O(N) generation time and O(1) history-state memory. For example, on challenging math problems, YANchor achieves 82.93% mean pass@1 on AIME 2024–2026 and 63.64% on HMMT, substantially outperforming linear-time, constant-state counterparts, including 7B–14B models. YANchor delivers several-fold higher batched long-generation throughput than Transformer and hybrid baselines on H100. Evaluations across dozens of benchmarks demonstrate its superiority in general-purpose capabilities.…