AL

ALMA 7B

AlmaGGUF by TheBlokeUpdated 2023-10-14
Download on Hugging Face
Parameters6.74B
Context4K
Layers32
Architecturellama
Smallest quant2.8 GB
Quick answer

ALMA 7B needs about 8.9 GB of memory at Q4_K_M (a 4.1 GB file) with an 8K context window, so it fits on a 12 GB GPU. This repository has 12 quants, from 2.8 GB (Q2_K) to 7.2 GB (Q8_0).

ALMA 7B – a 6.74B parameter translation model in GGUF format

ALMA 7B is a 6.74B parameter LLaMA‑based translation model fine‑tuned on monolingual and parallel data. It supports 4096 token context and is released under the MIT license. The model is available in multiple GGUF quantisations for CPU and GPU inference.

Good at

  • High‑quality translation
  • Low‑latency inference on modest hardware
  • Support for multiple quantisation levels

Limits

  • No vision support
  • Requires 8‑k context for memory estimates
  • No explicit language list provided

Best for

  • Local translation services
  • Embedded translation in web UI
  • Low‑cost inference on edge devices

Who it’s for. Developers and researchers needing a lightweight translation model that can run on local hardware. The model is suitable for integration into web interfaces or local applications where GPU resources are limited.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q2_K2.8 GB 5.5 GB
Fits Get file →
Q3_K_S2.9 GB 5.6 GB
Fits Get file →
Q3_K_M3.3 GB 5.9 GB
Fits Get file →
Q3_K_L3.6 GB 6.2 GB
Fits Get file →
Q4_03.8 GB 6.5 GB
Fits Get file →
Q4_K_S3.9 GB 6.5 GB
Fits Get file →
Q4_K_M4.1 GB 6.7 GB
Fits Get file →
Q5_04.7 GB 7.3 GB
Tight Get file →
Q5_K_S4.7 GB 7.3 GB
Tight Get file →
Q5_K_M4.8 GB 7.4 GB
Tight Get file →
Q6_K5.5 GB 8.2 GB
Too big Get file →
Q8_07.2 GB 9.8 GB
Too big Get file →

For 8 GB VRAM, use Q4_K_M (4.08 GB file, ~8.9 GB memory). For 12 GB VRAM, Q5_K_M (4.78 GB file, ~9.6 GB memory) is a good balance. If memory is very tight, Q3_K_S (2.95 GB file, ~7.7 GB memory) can be used.

What GPU do you need for ALMA 7B?

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: no quant fits at an 8K context.
  • 8 GB VRAM: the best quant that fits is Q3_K_S (2.9 GB file, about 7.7 GB in use).
  • 12 GB VRAM: the best quant that fits is Q8_0 (7.2 GB file, about 12.0 GB in use).

Which ALMA 7B quant should you choose?

Q4_K_M is the usual starting point for ALMA 7B: it balances file size and quality well. Here it is a 4.1 GB file.

The Q8_0 file is 7.2 GB, about 1.8 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 Q2_K at 2.8 GB, with the quality trade-offs that come with fewer bits. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturellama
Parameters6.74B
Layers32
Trained context4,096 tokens
KV cache (fp16)512 KB per token

Source

Quantized byTheBloke
FamilyAlma
Licensemit
Downloads (HF)764
Last updated2023-10-14

Frequently asked questions about ALMA 7B

How much VRAM does ALMA 7B need?

About 8.9 GB at Q4_K_M with an 8K context window. That is the 4.1 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 ALMA 7B on an 8 GB GPU?

Yes, with Q3_K_S: the 2.9 GB file needs about 7.7 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of ALMA 7B GGUF?

It depends on the quant: Q2_K is 2.8 GB, Q3_K_S is 2.9 GB, Q3_K_M is 3.3 GB, Q3_K_L is 3.6 GB, Q4_0 is 3.8 GB, Q4_K_S is 3.9 GB and 6 more.

What context length does ALMA 7B support?

The GGUF header lists a trained context length of 4,096 tokens (4K). The KV cache grows with the context you use: about 4.3 GB at 8K and 2.1 GB at 4K.

Where can I download ALMA 7B GGUF?

The files are on Hugging Face in the TheBloke/ALMA-7B-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 ALMA 7B use?

Hugging Face lists the license as mit. 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 ALMA 7B on 8 GB VRAM?

Yes, using Q4_K_M or Q3_K_S quantisations will fit within 8 GB VRAM with approximate memory usage of 8.9 GB or 7.7 GB respectively.

Does ALMA 7B support vision tasks?

No, the model is purely a language model and does not include vision capabilities.

What license applies to ALMA 7B?

The model is released under the MIT license, allowing free use and modification.

Similar models

Browse all →

More from TheBloke

All by TheBloke →

More Alma models

All Alma →

Other small models

All small models →