Shadow-Siren-14B-v2-Pruned is an open-weight model from Remesh, released under Apache License 2.0. It has 14.4B parameters and a 262,144-token context. At 16-bit it needs about 34.5 GB of GPU memory, which fits on 1x MI300X from $1.85 an hour, at the lowest prices in the SAVRN Index.
Surgically pruned 64-expert (~14.5B total) variant of Vortex5/Shadow-Siren-26B-A4B. - Sliced from 128 experts down to 64 experts across all MoE blocks. - Router gating matrix truncated cleanly to 64 rows.
Runs On
What it takes to serve Shadow-Siren-14B-v2-Pruned (14.4B 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 | 28.8 GB | 34.5 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 8-bit | 14.4 GB | 17.3 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 4-bit | 7.2 GB | 8.6 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 22, 2026.
Shadow-Siren-14B-v2-Pruned on every accelerator the SAVRN Index prices, at every precision
Model Card
By Remesh, published under apache-2.0, revision 2010de91e83f.
Surgically pruned 64-expert (~14.5B total) variant of Vortex5/Shadow-Siren-26B-A4B. - Sliced from 128 experts down to 64 experts across all MoE blocks. - Router gating matrix truncated cleanly to 64 rows. - Preserves original chat templates and tokenizer metadata for persistent multi-turn reasoning. - Full BF16 safetensors weights ready for fine-tuning or custom quantization.
Read Remesh's full model card
Shadow-Siren-14B-Pruned
Surgically pruned 64-expert (~14.5B total) variant of Vortex5/Shadow-Siren-26B-A4B.
Highlights
- Sliced from 128 experts down to 64 experts across all MoE blocks.
- Router gating matrix truncated cleanly to 64 rows.
- Preserves original chat templates and tokenizer metadata for persistent multi-turn reasoning.
- Full BF16 safetensors weights ready for fine-tuning or custom quantization.
Upstream Credits
- Base model: Vortex5/Shadow-Siren-26B-A4B
Configuration
- Architecture
- Gemma4ForConditionalGeneration
- Context length (tokens)
- 262,144
- Layers
- 30
- Hidden size
- 2,816
- Feed-forward size
- 2,112
- Attention heads
- 16
- Key/value heads
- 8
- Head dimension
- 256
- Vocabulary size
- 262,144
- Experts
- 64
- Sliding window (tokens)
- 1,024
- Model type
- gemma4
Identity and Version
- Repository
- Goopr/Shadow-Siren-14B-v2-Pruned
- Publisher
- Remesh
- Task
- Not stated by the source
- Modality
- Other
- Library
- Not stated by the source
- Parameters
- 14.4B parameters
- Languages
- en
- Revision
- 2010de91e83f3399166bb8dfc57419dec539e390
- First published
- 2026-10-03
- Last updated
- 2026-10-03
Files and Weights
19 files, 28.8 GB in total. The weights are 11 files totalling 28.8 GB in safetensors.
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| model-00001-of-00011.safetensors | Weights | 3.2 GB | f250f4fa11f3 |
| model-00002-of-00011.safetensors | Weights | 2.6 GB | fdb4e0d2f7d0 |
| model-00003-of-00011.safetensors | Weights | 2.6 GB | c7899f0aa966 |
| model-00004-of-00011.safetensors | Weights | 2.6 GB | 7c5aa7f1bf8b |
| model-00005-of-00011.safetensors | Weights | 2.6 GB | ad67d12ea912 |
| model-00006-of-00011.safetensors | Weights | 2.6 GB | b33a03125d9f |
| model-00007-of-00011.safetensors | Weights | 2.6 GB | 001d8b91d557 |
| model-00008-of-00011.safetensors | Weights | 2.6 GB | 09b39df4ec3b |
| model-00009-of-00011.safetensors | Weights | 2.6 GB | 16198ca03b04 |
| model-00010-of-00011.safetensors | Weights | 2.6 GB | a60ff9d9481c |
| model-00011-of-00011.safetensors | Weights | 2.0 GB | d64cc7d0e03a |
| config.json | Configuration | 3.8 KB | — |
| model.safetensors.index.json | Configuration | 103.2 KB | — |
| processor_config.json | Configuration | 1.7 KB | — |
| README.md | Documentation | 735 B | — |
| chat_template.jinja | Other | 18.7 KB | — |
| .gitattributes | Repository | 1.6 KB | — |
| tokenizer.json | Tokenizer | 32.2 MB | cc8d3a0ce364 |
| tokenizer_config.json | Tokenizer | 3.6 KB | — |
License and Download
- License
- apache-2.0
- Access
- Open weights, no gate
- Download size
- 28.8 GB
Released by Remesh through its official repository on Hugging Face. Read the license.
Built From
- Derived from Vortex5/Shadow-Siren-26B-A4B
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 28.8 GB |
| 16-bit | 28.8 GB |
| 8-bit | 14.4 GB |
| 4-bit | 7.2 GB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.
Questions About Shadow-Siren-14B-v2-Pruned
How much GPU memory does Shadow-Siren-14B-v2-Pruned need?
About 34.5 GB at 16-bit and 8.6 GB at 4-bit: the weights (14.4B parameters) plus a working margin. A long context needs more.
What is the cheapest GPU to run Shadow-Siren-14B-v2-Pruned 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 Shadow-Siren-14B-v2-Pruned commercially?
Yes. Shadow-Siren-14B-v2-Pruned 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 Shadow-Siren-14B-v2-Pruned's context length?
262,144 tokens, from the maximum position embeddings in its published configuration.