Open-weight model
hybrid-moe-30b-a3b-base-i1-GGUF
by Team Mradermacher mradermacher/hybrid-moe-30b-a3b-base-i1-GGUF
weighted/imatrix quants of https://huggingface.co/geodesic-research/hybrid-moe-30b-a3b-base For a convenient overview and download list, visit our model page for this model.
Model Card
weighted/imatrix quants of https://huggingface.co/geodesic-research/hybrid-moe-30b-a3b-base For a convenient overview and download list, visit our model page for this model. static quants are available at https://huggingface.co/mradermacher/hybrid-moe-30b-a3b-base-GGUF If you are unsure how to use GGUF files, refer to one of TheBloke's READMEs for more details, including on how to concatenate multi-part files. (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants) Here is a handy graph by ikawrakow comparing some lower-quality quant And here are Artefact2's thoughts on the matter…
Excerpt from the card by Team Mradermacher.
Identity and Version
- Repository
- mradermacher/hybrid-moe-30b-a3b-base-i1-GGUF
- Publisher
- Team Mradermacher
- Task
- Not stated by the source
- Modality
- Other
- Library
- transformers
- Parameters
- Not stated by the source
- Languages
- en
- Revision
- 80b7018ed80b1aceec4007bebd2316e48aee504a
- First published
- 2026-09-17
- Last updated
- 2026-09-18
Files and Weights
26 files, 459.5 GB in total. The weights are 24 files totalling 459.5 GB in gguf.
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| hybrid-moe-30b-a3b-base.i1-IQ1_M.gguf | Weights | 17.9 GB | 8258609527c7 |
| hybrid-moe-30b-a3b-base.i1-IQ1_S.gguf | Weights | 17.9 GB | 6ce400b6e23e |
| hybrid-moe-30b-a3b-base.i1-IQ2_M.gguf | Weights | 17.9 GB | 69d631ef686f |
| hybrid-moe-30b-a3b-base.i1-IQ2_S.gguf | Weights | 17.9 GB | 55e56783842d |
| hybrid-moe-30b-a3b-base.i1-IQ2_XS.gguf | Weights | 17.9 GB | c32249667b86 |
| hybrid-moe-30b-a3b-base.i1-IQ2_XXS.gguf | Weights | 17.9 GB | 126750b22bde |
| hybrid-moe-30b-a3b-base.i1-IQ3_M.gguf | Weights | 18.1 GB | 582732954f17 |
| hybrid-moe-30b-a3b-base.i1-IQ3_S.gguf | Weights | 17.9 GB | 815a08563947 |
| hybrid-moe-30b-a3b-base.i1-IQ3_XS.gguf | Weights | 17.9 GB | 4a294bbc85bd |
| hybrid-moe-30b-a3b-base.i1-IQ3_XXS.gguf | Weights | 17.9 GB | 1e14b566f5ae |
| hybrid-moe-30b-a3b-base.i1-IQ4_XS.gguf | Weights | 18.0 GB | 418b8f351ef9 |
| hybrid-moe-30b-a3b-base.i1-Q2_K.gguf | Weights | 17.9 GB | 8918de288eb6 |
| hybrid-moe-30b-a3b-base.i1-Q2_K_S.gguf | Weights | 18.1 GB | a89359ec348c |
| hybrid-moe-30b-a3b-base.i1-Q3_K_L.gguf | Weights | 20.7 GB | 3179f829f566 |
| hybrid-moe-30b-a3b-base.i1-Q3_K_M.gguf | Weights | 19.8 GB | d2efbfe6e7d7 |
| hybrid-moe-30b-a3b-base.i1-Q3_K_S.gguf | Weights | 17.9 GB | f2b22a16bce3 |
| hybrid-moe-30b-a3b-base.i1-Q4_0.gguf | Weights | 18.1 GB | 1c536101493f |
| hybrid-moe-30b-a3b-base.i1-Q4_1.gguf | Weights | 19.9 GB | 73dced3a2114 |
| hybrid-moe-30b-a3b-base.i1-Q4_K_M.gguf | Weights | 24.5 GB | 3a527cdf8ed4 |
| hybrid-moe-30b-a3b-base.i1-Q4_K_S.gguf | Weights | 21.9 GB | b56316646b20 |
| hybrid-moe-30b-a3b-base.i1-Q5_K_M.gguf | Weights | 26.0 GB | f277d82b06f9 |
| hybrid-moe-30b-a3b-base.i1-Q5_K_S.gguf | Weights | 23.8 GB | 5baf2737aba9 |
| hybrid-moe-30b-a3b-base.i1-Q6_K.gguf | Weights | 33.5 GB | b2113e456ad4 |
| hybrid-moe-30b-a3b-base.imatrix.gguf | Weights | 55.3 MB | 5bd9151264a2 |
| README.md | Documentation | 6.2 KB | — |
| .gitattributes | Repository | 3.3 KB | — |
License and Download
- License
- Not stated by the source
- Access
- Open weights, no gate
- Download size
- 459.5 GB
Released by Team Mradermacher through its official repository on Hugging Face.
Built From
- Derived from geodesic-research/hybrid-moe-30b-a3b-base
- Quantized from geodesic-research/hybrid-moe-30b-a3b-base
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 459.5 GB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.