LI

Ling 3.0 Flash

LingGGUF by AtomicChatUpdated 2026-10-28
Download on Hugging Face
Parameters124B
Context128K
Layers42
Architecturebailingmoe3
Smallest quant0.5 GB
Quick answer

Ling 3.0 Flash needs about 99.7 GB of memory at Q4_K_M (a 75.3 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 31 quants, from 0.5 GB to 248.9 GB (BF16). Longer context needs more: at 32K tokens the KV cache alone adds about 95.8 GB.

Ling 3.0 Flash – 124B MoE model with 5.1B active parameters

Ling 3.0 Flash is a 124B parameter model with 5.1B active parameters. It uses a native hybrid linear attention architecture and 512‑expert MoE. The model is released under the MIT license and is available in multiple GGUF quantizations.

Good at

  • Long‑context reasoning
  • Agentic tool calling
  • Efficient inference with MoE

Limits

  • Requires TurboQuant build
  • No vision support
  • Large memory footprint for high‑precision quants

Best for

  • Production agent workflows
  • Long‑horizon code generation
  • Complex reasoning tasks

Who it’s for. Developers and researchers needing a high‑capacity, efficient LLM for production. The model is suitable for local deployment with llama.cpp and requires a compatible GPU or CPU with sufficient VRAM.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
—0.5 GB 24.9 GB
Too big Get file →
IQ1_S32.4 GB 56.8 GB
Too big Get file →
IQ1_M36.5 GB 60.9 GB
Too big Get file →
IQ2_XXS39.2 GB 63.7 GB
Too big Get file →
IQ2_XS44.7 GB 69.2 GB
Too big Get file →
IQ2_S46.9 GB (2 parts) 71.4 GB
Too big Get file →
IQ2_M49.1 GB (2 parts) 73.6 GB
Too big Get file →
IQ3_XXS57.1 GB (2 parts) 81.5 GB
Too big Get file →
IQ3_S57.8 GB (2 parts) 82.3 GB
Too big Get file →
IQ3_M62.2 GB (2 parts) 86.6 GB
Too big Get file →
IQ4_XS66.4 GB (2 parts) 90.9 GB
Too big Get file →
IQ4_XS68.6 GB (2 parts) 93.0 GB
Too big Get file →
IQ4_XXS69.3 GB (2 parts) 93.8 GB
Too big Get file →
—72.3 GB (2 parts) 96.8 GB
Too big Get file →
—72.3 GB (2 parts) 96.8 GB
Too big Get file →
—72.3 GB (2 parts) 96.8 GB
Too big Get file →
Q4_K72.3 GB (2 parts) 96.8 GB
Too big Get file →
Q4_K_S74.2 GB (2 parts) 98.7 GB
Too big Get file →
IQ4_XS74.9 GB (2 parts) 99.3 GB
Too big Get file →
Q4_K_M75.3 GB (2 parts) 99.7 GB
Too big Get file →
IQ4_NL79.3 GB (2 parts) 103.8 GB
Too big Get file →
Q4_K_M79.3 GB (2 parts) 103.8 GB
Too big Get file →
Q4_K_L84.0 GB (2 parts) 108.4 GB
Too big Get file →
Q5_K_S87.4 GB (2 parts) 111.9 GB
Too big Get file →
Q5_K_M88.3 GB (2 parts) 112.8 GB
Too big Get file →
Q5_K_M89.4 GB (2 parts) 113.9 GB
Too big Get file →
Q5_K_L95.3 GB (3 parts) 119.8 GB
Too big Get file →
Q6_K102.2 GB (3 parts) 126.7 GB
Too big Get file →
Q6_K107.5 GB (3 parts) 132.0 GB
Too big Get file →
Q8_0133.1 GB (3 parts) 157.6 GB
Too big Get file →
BF16248.9 GB (6 parts) 273.4 GB
Too big Get file →

For 8–12 GB VRAM, use IQ1_S or IQ1_M (≈56–61 GB memory at 8k context). For 16–24 GB VRAM, IQ2_S or IQ2_M (≈69–73 GB). For 32 GB VRAM or more, Q4_K or Q5_K variants provide a balance of size and performance.

What GPU do you need for Ling 3.0 Flash?

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 the main file (0.5 GB file, about 24.9 GB in use).
  • 48 GB VRAM: the best quant that fits is the main file (0.5 GB file, about 24.9 GB in use).

Which Ling 3.0 Flash quant should you choose?

Q4_K_M is the usual starting point for Ling 3.0 Flash: it balances file size and quality well. Here it is a 75.3 GB file.

The Q8_0 file is 133.1 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 the first one in the table at 0.5 GB, with the quality trade-offs that come with fewer bits.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturebailingmoe3
Parameters124B (approximate, mixture of experts)
Layers42
Trained context131,072 tokens
KV cache (fp16)2856 KB per token

Source

Quantized byAtomicChat
FamilyLing
Licensemit
Downloads (HF)230,585
Last updated2026-10-28

Frequently asked questions about Ling 3.0 Flash

How much VRAM does Ling 3.0 Flash need?

About 99.7 GB at Q4_K_M with an 8K context window. That is the 75.3 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 Ling 3.0 Flash on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 24.9 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Ling 3.0 Flash GGUF?

It depends on the quant: file is 0.5 GB, IQ1_S is 32.4 GB, IQ1_M is 36.5 GB, IQ2_XXS is 39.2 GB, IQ2_XS is 44.7 GB, IQ2_S is 46.9 GB and 25 more.

What context length does Ling 3.0 Flash support?

The GGUF header lists a trained context length of 131,072 tokens (128K). The KV cache grows with the context you use: about 24.0 GB at 8K and 95.8 GB at 32K.

Where can I download Ling 3.0 Flash GGUF?

The files are on Hugging Face in the AtomicChat/Ling-3.0-flash-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 Ling 3.0 Flash 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 Ling 3.0 Flash on 8 GB VRAM?

No, the smallest quant requires at least 24.9 GB memory at 8k context. 8 GB VRAM is insufficient.

Does Ling 3.0 Flash support vision?

No, the model has no vision capabilities.

Other GGUFs of Ling-3.0-flash

Similar models

Browse all →

More from AtomicChat

All by AtomicChat →

More Ling models

All Ling →

Other large models

All large models →