FI

Finch 3b

FinchGGUF by latestissueUpdated 2024-10-24
Download on Hugging Face
Parameters3.1B
Context1024K
Layers32
Architecturerwkv6
Smallest quant1.2 GB
Quick answer

Finch 3b needs about 2.6 GB of memory at Q4_K (a 1.9 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 23 quants, from 1.2 GB (Q2_K) to 12.4 GB (F32).

Finch 3b – a 3 billion‑parameter RWKV‑6 model

Finch 3b is a 3 billion‑parameter RWKV‑6 model released under the Apache‑2.0 license. It supports a maximum context length of 1 048 576 tokens. The model is available in multiple GGUF quantizations ranging from Q2_K to F32.

Good at

  • Large context handling
  • Low‑memory inference
  • Versatile token generation

Limits

  • No vision support
  • Requires 1 048 576 token context for full training size
  • License is Apache‑2.0

Best for

  • Long‑form text generation
  • Context‑heavy dialogue systems
  • Research prototyping

Who it’s for. Developers and researchers needing a lightweight RWKV model with extensive context. The model is suitable for local deployment on systems with moderate GPU memory.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q2_K1.2 GB 1.8 GB
Fits Get file →
IQ3_XXS1.3 GB 2.0 GB
Fits Get file →
IQ3_S1.5 GB 2.2 GB
Fits Get file →
Q3_K1.5 GB 2.2 GB
Fits Get file →
Q3_K_L1.5 GB 2.2 GB
Fits Get file →
IQ4_XS1.8 GB 2.5 GB
Fits Get file →
IQ4_NL1.9 GB 2.6 GB
Fits Get file →
Q4_01.9 GB 2.6 GB
Fits Get file →
Q4_01.9 GB 2.6 GB
Fits Get file →
Q4_01.9 GB 2.6 GB
Fits Get file →
Q4_01.9 GB 2.6 GB
Fits Get file →
Q4_K1.9 GB 2.6 GB
Fits Get file →
Q4_K_S1.9 GB 2.6 GB
Fits Get file →
Q4_12.1 GB 2.8 GB
Fits Get file →
Q5_02.2 GB 3.0 GB
Fits Get file →
Q5_K2.2 GB 3.0 GB
Fits Get file →
Q5_K_S2.2 GB 3.0 GB
Fits Get file →
Q5_12.4 GB 3.2 GB
Fits Get file →
Q6_K2.6 GB 3.4 GB
Fits Get file →
Q8_03.4 GB 4.3 GB
Fits Get file →
BF166.3 GB 7.5 GB
Tight Get file →
F166.3 GB 7.5 GB
Tight Get file →
F3212.4 GB 14.4 GB
Too big Get file →

For 8 GB VRAM, use Q4_0 (1.88 GB file, ~2.6 GB memory). For 12 GB VRAM, Q5_0 (2.25 GB file, ~3 GB memory) is a good balance. For 16 GB VRAM, Q6_K (2.63 GB file, ~3.4 GB memory) offers higher precision.

What GPU do you need for Finch 3b?

These figures add the file size, an fp16 KV cache at an 8K context window 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: the best quant that fits is Q6_K (2.6 GB file, about 3.4 GB in use).
  • 6 GB VRAM: the best quant that fits is Q8_0 (3.4 GB file, about 4.3 GB in use).
  • 8 GB VRAM: the best quant that fits is F16 (6.3 GB file, about 7.5 GB in use).
  • 12 GB VRAM: the best quant that fits is F16 (6.3 GB file, about 7.5 GB in use).
  • 16 GB VRAM: the best quant that fits is F32 (12.4 GB file, about 14.4 GB in use).

Which Finch 3b quant should you choose?

Q4_K (1.9 GB) is the middle option in this repository. The usual sweet spot between file size and quality, and the quant most people start with.

If memory is tight, the smallest file is Q2_K at 1.2 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

Architecturerwkv6
Parameters3.1B
Layers32
Trained context1,048,576 tokens

Source

Quantized bylatestissue
FamilyFinch
Licenseapache-2.0
Downloads (HF)6,812
Last updated2024-10-24

Frequently asked questions about Finch 3b

How much VRAM does Finch 3b need?

About 2.6 GB at Q4_K with an 8K context window. That is the 1.9 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 Finch 3b on an 8 GB GPU?

Yes, with F16: the 6.3 GB file needs about 7.5 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of Finch 3b GGUF?

It depends on the quant: Q2_K is 1.2 GB, IQ3_XXS is 1.3 GB, IQ3_S is 1.5 GB, Q3_K is 1.5 GB, Q3_K_L is 1.5 GB, IQ4_XS is 1.8 GB and 17 more.

What context length does Finch 3b support?

The GGUF header lists a trained context length of 1,048,576 tokens (1024K).

Where can I download Finch 3b GGUF?

The files are on Hugging Face in the latestissue/rwkv-6-finch-3b-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 Finch 3b 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 Finch 3b on 8 GB VRAM?

Yes, using the Q4_0 quantization which requires about 2.6 GB of VRAM at 8 k context.

What is the maximum context length?

The model was trained with a context of 1 048 576 tokens.

Is Finch 3b licensed for commercial use?

It is released under the Apache‑2.0 license, which permits commercial use.

Similar models

Browse all →

More from latestissue

All by latestissue →

More Finch models

All Finch →

Other tiny models

All tiny models →