AL

Source Hf Aliceai T5 35b A0.6B

AliceaiGGUF by ngquocvinhUpdated 2026-09-23
Download on Hugging Face
Parameters34.6B
Context128K
Layers16
Architecturealiceai_t5_moe
Smallest quant5.1 GB
Quick answer

Source Hf Aliceai T5 35b A0.6B needs about 21.8 GB of memory at Q4_K_M (a 21.0 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 26 quants, from 5.1 GB (Q1_0) to 69.2 GB (BF16). Longer context needs more: at 32K tokens the KV cache alone adds about 1.1 GB.

AliceAI‑T5 35B encoder‑decoder with sparse MoE

Source Hf Aliceai T5 35b A0.6B is a 35 billion‑parameter encoder‑decoder model with 0.6 billion active MoE parameters. It supports 131 073‑token context and is licensed under Apache‑2.0. The model is available in multiple GGUF quantizations.

Good at

  • Coding tasks
  • Encoder‑decoder generation
  • Sparse MoE inference

Limits

  • No vision support
  • Requires AliceAI‑aware llama.cpp runtime
  • Context limited to 131 073 tokens

Best for

  • Generating code snippets
  • Translating long documents
  • Fine‑tuning with LoRA

Who it’s for. Developers needing a large encoder‑decoder model for code or text generation. The model is suitable for research and production workloads that can run on a custom AliceAI‑aware runtime.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q1_05.1 GB 5.9 GB
Fits Get file →
IQ1_S7.1 GB 7.9 GB
Tight Get file →
TQ1_07.6 GB 8.4 GB
Too big Get file →
TQ2_09.2 GB 10.0 GB
Too big Get file →
Q2_010.0 GB 10.8 GB
Too big Get file →
IQ2_S10.3 GB 11.1 GB
Too big Get file →
Q2_K_S11.8 GB 12.6 GB
Too big Get file →
Q2_K12.7 GB 13.5 GB
Too big Get file →
IQ3_XXS13.4 GB 14.2 GB
Too big Get file →
Q3_K_S15.0 GB 15.8 GB
Too big Get file →
Q3_K_M16.6 GB 17.3 GB
Too big Get file →
Q3_K_L17.9 GB 18.7 GB
Too big Get file →
IQ4_XS18.5 GB 19.3 GB
Too big Get file →
MXFP4_MOE18.8 GB 19.6 GB
Too big Get file →
IQ4_NL19.6 GB 20.3 GB
Too big Get file →
Q4_019.7 GB 20.4 GB
Too big Get file →
Q4_K_S19.7 GB 20.5 GB
Too big Get file →
Q4_K_M21.0 GB 21.8 GB
Too big Get file →
Q4_121.7 GB 22.5 GB
Too big Get file →
Q5_K_S23.9 GB 24.6 GB
Too big Get file →
Q5_023.9 GB 24.7 GB
Too big Get file →
Q5_K_M24.6 GB 25.4 GB
Too big Get file →
Q5_126.0 GB 26.8 GB
Too big Get file →
Q6_K28.4 GB 29.2 GB
Too big Get file →
Q8_036.8 GB 37.6 GB
Too big Get file →
BF1669.2 GB 69.9 GB
Too big Get file →

For 8 GB VRAM, use Q4_0 (≈20.4 GB memory at 8k context). For 16 GB VRAM, Q4_K_M (≈21.8 GB). For 32 GB VRAM, Q5_K_M (≈25.4 GB). Memory estimates are approximate.

What GPU do you need for Source Hf Aliceai T5 35b A0.6B?

These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.

  • 4 GB VRAM: no quant fits at an 8K context.
  • 6 GB VRAM: the best quant that fits is Q1_0 (5.1 GB file, about 5.9 GB in use).
  • 8 GB VRAM: the best quant that fits is IQ1_S (7.1 GB file, about 7.9 GB in use).
  • 12 GB VRAM: the best quant that fits is IQ2_S (10.3 GB file, about 11.1 GB in use).
  • 16 GB VRAM: the best quant that fits is Q3_K_S (15.0 GB file, about 15.8 GB in use).
  • 24 GB VRAM: the best quant that fits is Q4_1 (21.7 GB file, about 22.5 GB in use).
  • 32 GB VRAM: the best quant that fits is Q6_K (28.4 GB file, about 29.2 GB in use).
  • 48 GB VRAM: the best quant that fits is Q8_0 (36.8 GB file, about 37.6 GB in use).

Which Source Hf Aliceai T5 35b A0.6B quant should you choose?

Q4_K_M is the usual starting point for Source Hf Aliceai T5 35b A0.6B: it balances file size and quality well. Here it is a 21.0 GB file.

The Q8_0 file is 36.8 GB, about 1.7 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.

If memory is tight, the smallest file is Q1_0 at 5.1 GB, with the quality trade-offs that come with fewer bits. Lower bit counts mean smaller files and more quality loss.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturealiceai_t5_moe
Parameters34.6B (approximate, mixture of experts)
Active parameters0.6B
Layers16
Trained context131,073 tokens
KV cache (fp16)32 KB per token

Source

Quantized byngquocvinh
FamilyAliceai
Licenseapache-2.0
Downloads (HF)14,245
Last updated2026-09-23

Frequently asked questions about Source Hf Aliceai T5 35b A0.6B

How much VRAM does Source Hf Aliceai T5 35b A0.6B need?

About 21.8 GB at Q4_K_M with an 8K context window. That is the 21.0 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.

Can I run Source Hf Aliceai T5 35b A0.6B on an 8 GB GPU?

Yes, with IQ1_S: the 7.1 GB file needs about 7.9 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of Source Hf Aliceai T5 35b A0.6B GGUF?

It depends on the quant: Q1_0 is 5.1 GB, IQ1_S is 7.1 GB, TQ1_0 is 7.6 GB, TQ2_0 is 9.2 GB, Q2_0 is 10.0 GB, IQ2_S is 10.3 GB and 20 more.

What context length does Source Hf Aliceai T5 35b A0.6B support?

The GGUF header lists a trained context length of 131,073 tokens (128K). The KV cache grows with the context you use: about 0.3 GB at 8K and 1.1 GB at 32K.

Where can I download Source Hf Aliceai T5 35b A0.6B GGUF?

The files are on Hugging Face in the ngquocvinh/AliceAI-T5-35B-A0.6B-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does Source Hf Aliceai T5 35b A0.6B use?

Hugging Face lists the license as apache-2.0. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.

Can I run this model on 8 GB VRAM?

Use the Q4_0 quantization; it requires about 20.4 GB of VRAM at 8k context, so 8 GB is insufficient.

Does this model support image input?

No, the model has no vision capabilities.

What runtime is needed?

An AliceAI‑aware llama.cpp runtime with encoder, cross‑attention, and decoder support is required.

Other GGUFs of AliceAI-T5-35B-A0.6B

Similar models

Browse all →

More from ngquocvinh

All by ngquocvinh →

More Aliceai models

All Aliceai →

Other large models

All large models →