LZ

lzlv longLORA 70b rope8 32k fp16

LzlvGGUF by mradermacherUpdated 2024-05-06
Download on Hugging Face
Parameters69B
Context4K
Layers80
Architecturellama
Smallest quant25.8 GB
Quick answer

lzlv longLORA 70b rope8 32k fp16 needs about 44.9 GB of memory at Q4_K_M (a 41.7 GB file) with an 8K context window, so it fits on a 48 GB GPU. This repository has 12 quants, from 25.8 GB (Q2_K) to 49.1 GB (Q5_K_M).

Lzlv longLORA 70B with 32k context, fp16

The lzlv longLORA 70b rope8 32k fp16 model merges LongAlpaca‑70B‑lora into lzlv_70b_fp16_hf. It supports a 32k token context using linear rope scaling 8. The model is 69 billion parameters and is available in multiple GGUF quantizations.

Good at

  • Long‑context generation
  • High‑parameter LLM tasks
  • Fine‑tuned Lzlv architecture

Limits

  • No vision support
  • CC‑BY‑NC‑2.0 license
  • Context limited to 4096 tokens for training

Best for

  • Research prototypes needing 32k context
  • Custom fine‑tuning on long sequences
  • Testing quantization trade‑offs

Who it’s for. Engineers and researchers who need a large‑parameter model with extended context. The model is suitable for experimentation and custom integration rather than production deployment.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q2_K25.8 GB 27.6 GB
Too big Get file →
IQ3_XS28.5 GB 30.3 GB
Too big Get file →
IQ3_S30.2 GB 32.1 GB
Too big Get file →
Q3_K_S30.2 GB 32.1 GB
Too big Get file →
IQ3_M31.3 GB 33.1 GB
Too big Get file →
Q3_K_M33.6 GB 35.4 GB
Too big Get file →
Q3_K_L36.5 GB 38.3 GB
Too big Get file →
IQ4_XS37.5 GB 39.3 GB
Too big Get file →
Q4_K_S39.6 GB 41.4 GB
Too big Get file →
Q4_K_M41.7 GB 43.6 GB
Too big Get file →
Q5_K_S47.8 GB 49.6 GB
Too big Get file →
Q5_K_M49.1 GB 50.9 GB
Too big Get file →

For 8 GB VRAM, use Q2_K (25.77 GB file, ~29 GB memory at 8k context). For 12 GB VRAM, Q3_K_S (30.23 GB file, ~33.4 GB memory). For 16 GB VRAM, Q4_K_M (41.73 GB file, ~44.9 GB memory). Memory estimates are approximate.

What GPU do you need for lzlv longLORA 70b rope8 32k fp16?

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: no quant fits at an 8K context.
  • 12 GB VRAM: no quant fits at an 8K context.
  • 16 GB VRAM: no quant fits at an 8K context.
  • 24 GB VRAM: no quant fits at an 8K context.
  • 32 GB VRAM: the best quant that fits is IQ3_XS (28.5 GB file, about 31.6 GB in use).
  • 48 GB VRAM: the best quant that fits is Q4_K_M (41.7 GB file, about 44.9 GB in use).

Which lzlv longLORA 70b rope8 32k fp16 quant should you choose?

Q4_K_M is the usual starting point for lzlv longLORA 70b rope8 32k fp16: it balances file size and quality well. Here it is a 41.7 GB file.

If memory is tight, the smallest file is Q2_K at 25.8 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
Parameters69B
Layers80
Trained context4,096 tokens
KV cache (fp16)320 KB per token

Source

Quantized bymradermacher
FamilyLzlv
Downloads (HF)672
Last updated2024-05-06

Frequently asked questions about lzlv longLORA 70b rope8 32k fp16

How much VRAM does lzlv longLORA 70b rope8 32k fp16 need?

About 44.9 GB at Q4_K_M with an 8K context window. That is the 41.7 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 lzlv longLORA 70b rope8 32k fp16 on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 29.0 GB, so you would need a larger GPU or CPU offloading.

What is the file size of lzlv longLORA 70b rope8 32k fp16 GGUF?

It depends on the quant: Q2_K is 25.8 GB, IQ3_XS is 28.5 GB, IQ3_S is 30.2 GB, Q3_K_S is 30.2 GB, IQ3_M is 31.3 GB, Q3_K_M is 33.6 GB and 6 more.

What context length does lzlv longLORA 70b rope8 32k fp16 support?

The GGUF header lists a trained context length of 4,096 tokens (4K). The KV cache grows with the context you use: about 2.7 GB at 8K and 1.3 GB at 4K.

Where can I download lzlv longLORA 70b rope8 32k fp16 GGUF?

The files are on Hugging Face in the mradermacher/lzlv-longLORA-70b-rope8-32k-fp16-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 lzlv longLORA 70b rope8 32k fp16 use?

Hugging Face lists the license as cc-by-nc-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 Q2_K quant; it requires about 29 GB of VRAM at 8k context, which exceeds 8 GB. A smaller quant is needed.

Does the model support vision tasks?

No, the model has no vision capabilities.

What license applies to this model?

The model is licensed under CC‑BY‑NC‑2.0.

Other GGUFs of lzlv-longLORA-70b-rope8-32k-fp16

Similar models

Browse all →

More from mradermacher

All by mradermacher →

More Lzlv models

All Lzlv →

Other large models

All large models →