OR

Orion-14B-Chat-0126-ck19800

OrionGGUF by OrionStarAIUpdated 2024-04-11
Download on Hugging Face
Parameters14.5B
Context4K
Layers40
Architectureorion
Smallest quant29.0 GB
Quick answer

Orion-14B-Chat-0126-ck19800 needs about 36.2 GB of memory at its main file (a 29.0 GB file) with an 8K context window, so it fits on a 48 GB GPU.

Orion-14B-Chat-0126-ck19800 – a 14.5B multilingual chat model

Orion-14B-Chat-0126-ck19800 is a 14.5B parameter model fine‑tuned for chat. It is trained on a 2.5T multilingual corpus and supports coding tasks. The model is available in a single quantized GGUF file.

Good at

  • Multilingual chat
  • Coding assistance
  • Large context handling

Limits

  • No vision support
  • License not specified
  • Requires large VRAM

Best for

  • Developers building chatbots
  • Researchers testing multilingual models
  • Code generation tools

Who it’s for. The model is suitable for developers and researchers who need a multilingual chat model with coding capabilities. It is ideal for projects that can accommodate the memory requirements of a 14.5B parameter model.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
—29.0 GB 32.9 GB
Too big Get file →

The only available GGUF file is 29 GB and estimates 36.2 GB of VRAM at 8 k context. This is approximate. For systems with less VRAM, a lower‑bit quantization would be required, but no such file is provided.

What GPU do you need for Orion-14B-Chat-0126-ck19800?

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: no quant fits at an 8K context.
  • 48 GB VRAM: the best quant that fits is the main file (29.0 GB file, about 36.2 GB in use).

Specifications

Architecture from the GGUF header

Architectureorion
Parameters14.5B
Layers40
Trained context4,096 tokens
KV cache (fp16)800 KB per token

Source

Quantized byOrionStarAI
FamilyOrion
Downloads (HF)3,425
Last updated2024-04-11

Frequently asked questions about Orion-14B-Chat-0126-ck19800

How much VRAM does Orion-14B-Chat-0126-ck19800 need?

About 36.2 GB at its main file with an 8K context window. That is the 29.0 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 Orion-14B-Chat-0126-ck19800 on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 36.2 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Orion-14B-Chat-0126-ck19800 GGUF?

It depends on the quant: file is 29.0 GB.

What context length does Orion-14B-Chat-0126-ck19800 support?

The GGUF header lists a trained context length of 4,096 tokens (4K). The KV cache grows with the context you use: about 6.7 GB at 8K and 3.4 GB at 4K.

Where can I download Orion-14B-Chat-0126-ck19800 GGUF?

The files are on Hugging Face in the OrionStarAI/Orion-14B-Chat repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

Can I run Orion-14B-Chat-0126-ck19800 on 8 GB VRAM?

The model requires approximately 36.2 GB of VRAM at 8 k context, so 8 GB VRAM is insufficient.

Does the model support vision tasks?

No, the model does not support vision.

What languages does the model handle?

It is multilingual, trained on Chinese, English, Japanese, Korean, and other languages.

Similar models

Browse all →

More Orion models

All Orion →

Other medium models

All medium models →