WH

whale v3 base merged i1

WhaleGGUF by mradermacherUpdated 2025-07-11
Download on Hugging Face
Parameters37.4B
Context160K
Layers61
Architecturedeepseek2
Smallest quant8.7 GB
Quick answer

whale v3 base merged i1 needs about 24.0 GB of memory at Q4_K_M (a 22.4 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 23 quants, from 8.7 GB (IQ1_S) to 30.8 GB (Q6_K). Longer context needs more: at 32K tokens the KV cache alone adds about 4.3 GB.

37.4B parameter Deepseek2 model with tools support

whale v3 base merged i1 is a 37.4B parameter Deepseek2 model. It was trained on 163840 context tokens and includes mixture of experts. The model is available in multiple GGUF quantizations.

Good at

  • Tools integration
  • Large context handling
  • Deepseek2 architecture

Limits

  • No vision support
  • Requires at least 10.2GB VRAM for IQ1_S
  • License not specified

Best for

  • Tool‑driven chatbots
  • Long‑form content generation
  • Custom LLM deployments

Who it’s for. Developers needing a large‑parameter Deepseek2 model with tools capability. The model is useful for building custom chat or automation systems where a 163840 token context is required.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ1_S8.7 GB 10.2 GB
Too big Get file →
IQ1_M9.3 GB 10.9 GB
Too big Get file →
IQ2_XXS10.3 GB 11.9 GB
Too big Get file →
IQ2_XS11.4 GB 13.0 GB
Too big Get file →
IQ2_S12.6 GB 14.2 GB
Too big Get file →
Q2_K_S13.1 GB 14.7 GB
Too big Get file →
IQ2_M13.5 GB 15.0 GB
Too big Get file →
Q2_K14.4 GB 16.0 GB
Too big Get file →
IQ3_XXS15.1 GB 16.7 GB
Too big Get file →
IQ3_XS16.0 GB 17.5 GB
Too big Get file →
IQ3_S16.5 GB 18.1 GB
Too big Get file →
Q3_K_S16.5 GB 18.1 GB
Too big Get file →
IQ3_M17.6 GB 19.2 GB
Too big Get file →
Q3_K_M18.6 GB 20.2 GB
Too big Get file →
IQ4_XS20.2 GB 21.8 GB
Too big Get file →
Q3_K_L20.5 GB 22.0 GB
Too big Get file →
Q4_021.4 GB 23.0 GB
Too big Get file →
Q4_K_S21.5 GB 23.1 GB
Too big Get file →
Q4_K_M22.4 GB 24.0 GB
Too big Get file →
Q4_123.6 GB 25.2 GB
Too big Get file →
Q5_K_S25.9 GB 27.5 GB
Too big Get file →
Q5_K_M26.4 GB 28.0 GB
Too big Get file →
Q6_K30.8 GB 32.4 GB
Too big Get file →

For 8GB VRAM, the smallest quant IQ1_S (8.66GB) is still above the limit; the next smallest IQ1_M (9.29GB) also exceeds 8GB. For 10GB VRAM, IQ1_S (est. 10.2GB) is the closest fit. For 12GB VRAM, IQ1_M or IQ2_XXS (est. 10.9–11.9GB) are suitable. Larger VRAM allows higher‑quality quants such as IQ2_S or Q2_K.

What GPU do you need for whale v3 base merged i1?

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: the best quant that fits is IQ2_XXS (10.3 GB file, about 11.9 GB in use).
  • 16 GB VRAM: the best quant that fits is Q2_K (14.4 GB file, about 16.0 GB in use).
  • 24 GB VRAM: the best quant that fits is Q4_K_M (22.4 GB file, about 24.0 GB in use).
  • 32 GB VRAM: the best quant that fits is Q5_K_M (26.4 GB file, about 28.0 GB in use).
  • 48 GB VRAM: the best quant that fits is Q6_K (30.8 GB file, about 32.4 GB in use).

Which whale v3 base merged i1 quant should you choose?

Q4_K_M is the usual starting point for whale v3 base merged i1: it balances file size and quality well. Here it is a 22.4 GB file.

If memory is tight, the smallest file is IQ1_S at 8.7 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

Architecturedeepseek2
Parameters37.4B (approximate, mixture of experts)
Layers61
Trained context163,840 tokens
KV cache (fp16)130 KB per token

Source

Quantized bymradermacher
FamilyWhale
Downloads (HF)795
Last updated2025-07-11

Frequently asked questions about whale v3 base merged i1

How much VRAM does whale v3 base merged i1 need?

About 24.0 GB at Q4_K_M with an 8K context window. That is the 22.4 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 whale v3 base merged i1 on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 10.2 GB, so you would need a larger GPU or CPU offloading.

What is the file size of whale v3 base merged i1 GGUF?

It depends on the quant: IQ1_S is 8.7 GB, IQ1_M is 9.3 GB, IQ2_XXS is 10.3 GB, IQ2_XS is 11.4 GB, IQ2_S is 12.6 GB, Q2_K_S is 13.1 GB and 17 more.

What context length does whale v3 base merged i1 support?

The GGUF header lists a trained context length of 163,840 tokens (160K). The KV cache grows with the context you use: about 1.1 GB at 8K and 4.3 GB at 32K.

Where can I download whale v3 base merged i1 GGUF?

The files are on Hugging Face in the mradermacher/whale-v3-base-merged-i1-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

Can I run this model on 8GB VRAM?

No. The smallest quant IQ1_S requires about 10.2GB VRAM at 8k context, which exceeds 8GB.

What quant should I use for 10GB VRAM?

IQ1_S is the closest fit, estimating 10.2GB VRAM usage at 8k context.

Does this model support vision?

No, the model has no vision capability.

Other GGUFs of whale-v3-base-merged

Similar models

Browse all →

More from mradermacher

All by mradermacher →

More Whale models

All Whale →

Other large models

All large models →