OR

Whale 4B Thinking

OrcaGGUF by mradermacherUpdated 2026-09-26
Download on Hugging Face
Parameters4B
Context256K
Layers36
Architectureqwen3
Smallest quant0.0 GB
Quick answer

Whale 4B Thinking needs about 4.2 GB of memory at Q4_K_M (a 2.5 GB file) with an 8K context window, so it fits on a 6 GB GPU. This repository has 25 quants, from 0.0 GB to 3.3 GB (Q6_K). Longer context needs more: at 32K tokens the KV cache alone adds about 4.8 GB.

Whale 4B Thinking – a 4B‑parameter Qwen3 model specialized for deep chain‑of‑thought reasoning

Whale 4B Thinking is a 4.0B‑parameter causal language model built on Qwen3. It supports a native context length of 262144 tokens and is optimized for long, structured reasoning traces in the DeepSeek‑R1 style. The model is released under the apache‑2.0 license.

Good at

  • Deep chain‑of‑thought reasoning
  • Self‑verification and reflection
  • Long context handling

Limits

  • No vision support
  • Requires 8k context for memory estimates
  • Only available in GGUF format

Best for

  • Complex problem solving
  • Mathematics and code reasoning
  • Long‑form analytical tasks

Who it’s for. Developers and researchers needing a lightweight model that excels at reasoning tasks. The model is suitable for local deployment with llama.cpp, vLLM, or SGLang.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
—0.0 GB 1.7 GB
Fits Get file →
IQ1_S1.1 GB 2.8 GB
Fits Get file →
IQ1_M1.1 GB 2.8 GB
Fits Get file →
IQ2_XXS1.2 GB 3.0 GB
Fits Get file →
IQ2_XS1.4 GB 3.1 GB
Fits Get file →
IQ2_S1.4 GB 3.1 GB
Fits Get file →
IQ2_M1.5 GB 3.2 GB
Fits Get file →
Q2_K_S1.6 GB 3.3 GB
Fits Get file →
Q2_K1.7 GB 3.4 GB
Fits Get file →
IQ3_XXS1.7 GB 3.4 GB
Fits Get file →
IQ3_XS1.8 GB 3.5 GB
Fits Get file →
Q3_K_S1.9 GB 3.6 GB
Fits Get file →
IQ3_S1.9 GB 3.6 GB
Fits Get file →
IQ3_M2.0 GB 3.7 GB
Fits Get file →
Q3_K_M2.1 GB 3.8 GB
Fits Get file →
Q3_K_L2.2 GB 3.9 GB
Fits Get file →
IQ4_XS2.3 GB 4.0 GB
Fits Get file →
Q4_02.4 GB 4.1 GB
Fits Get file →
IQ4_NL2.4 GB 4.1 GB
Fits Get file →
Q4_K_S2.4 GB 4.1 GB
Fits Get file →
Q4_K_M2.5 GB 4.2 GB
Fits Get file →
Q4_12.6 GB 4.3 GB
Fits Get file →
Q5_K_S2.8 GB 4.5 GB
Fits Get file →
Q5_K_M2.9 GB 4.6 GB
Fits Get file →
Q6_K3.3 GB 5.0 GB
Fits Get file →

For 8 GB VRAM, use Q4_K_M (file size 2.08 GB, estimated memory 3.8 GB). For 12 GB VRAM, Q5_K_M (2.89 GB, 4.6 GB) or Q6_K (3.31 GB, 5 GB) provide higher quality. For 4 GB VRAM, IQ1_S (1.06 GB, 2.8 GB) or IQ2_XXS (1.25 GB, 3 GB) are suitable.

What GPU do you need for Whale 4B Thinking?

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: the best quant that fits is IQ4_XS (2.3 GB file, about 4.0 GB in use).
  • 6 GB VRAM: the best quant that fits is Q6_K (3.3 GB file, about 5.0 GB in use).

Which Whale 4B Thinking quant should you choose?

Q4_K_M is the usual starting point for Whale 4B Thinking: it balances file size and quality well. Here it is a 2.5 GB file.

If memory is tight, the smallest file is the first one in the table at 0.0 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

Architectureqwen3
Parameters4B
Layers36
Trained context262,144 tokens
KV cache (fp16)144 KB per token

Source

Quantized bymradermacher
FamilyOrca
Licenseapache-2.0
Downloads (HF)379
Last updated2026-09-26

Frequently asked questions about Whale 4B Thinking

How much VRAM does Whale 4B Thinking need?

About 4.2 GB at Q4_K_M with an 8K context window. That is the 2.5 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 4B Thinking on an 8 GB GPU?

Yes, with Q6_K: the 3.3 GB file needs about 5.0 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of Whale 4B Thinking GGUF?

It depends on the quant: file is 0.0 GB, IQ1_S is 1.1 GB, IQ1_M is 1.1 GB, IQ2_XXS is 1.2 GB, IQ2_XS is 1.4 GB, IQ2_S is 1.4 GB and 19 more.

What context length does Whale 4B Thinking support?

The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 1.2 GB at 8K and 4.8 GB at 32K.

Where can I download Whale 4B Thinking GGUF?

The files are on Hugging Face in the mradermacher/Whale-4B-Thinking-i1-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 Whale 4B Thinking 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 Whale 4B Thinking on 8 GB VRAM?

Yes, a Q4_K_M quant uses about 3.8 GB of VRAM at 8k context, which fits within an 8 GB GPU.

Does Whale 4B Thinking support vision or multimodal inputs?

No, the model is a purely text‑based causal language model with no vision capability.

Other GGUFs of Orca-4B-Thinking

Similar models

Browse all →

More from mradermacher

All by mradermacher →

More Orca models

All Orca →

Other small models

All small models →