TE

Tessera-27B-B-CJK

TesseraGGUF by ZelligeAIUpdated 2026-10-14
Download on Hugging Face
Parameters27.3B
Context256K
Layers65
Architectureqwen35
Smallest quant13.1 GB
Quick answer

Tessera-27B-B-CJK needs about 23.6 GB of memory at Q5_K_XL (a 20.9 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 6 quants, from 13.1 GB (Q3_K_XL) to 54.7 GB (BF16). Longer context needs more: at 32K tokens the KV cache alone adds about 8.7 GB.

Tessera-27B-B-CJK – a 27B coding‑agent model with compressed CJK thinking

Tessera-27B-B-CJK is a 27‑billion‑parameter coding‑agent built on Qwen3.8‑27B. It uses compressed CJK to reduce thinking tokens, while answers and tool calls remain in plain English and code. The model is released under Apache‑2.0 and supports vision input.

Good at

  • Compressed CJK thinking reduces token usage
  • Coding‑agent loops with reasoning_effort="xhigh"
  • Vision input via mmproj

Limits

  • Requires large VRAM for 8k context (≈15.8–34.1 GB depending on quant)
  • No safety tuning beyond base model
  • Not tested for image input

Best for

  • Coding‑agent loops with high reasoning effort
  • Short‑turn agent tasks with compressed CJK
  • Vision‑enabled coding queries

Who it’s for. Developers building coding‑assistant applications that need efficient token usage and can handle large VRAM requirements. The model is suitable for users who can run llama.cpp with the provided GGUF files and have access to a GPU with sufficient memory.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q3_K_XL13.1 GB 15.8 GB
Too big Get file →
Q4_K_XL17.6 GB 20.2 GB
Too big Get file →
Q5_K_XL20.9 GB 23.6 GB
Too big Get file →
Q6_K_XL25.3 GB 28.0 GB
Too big Get file →
Q8_K_XL31.5 GB 34.1 GB
Too big Get file →
BF1654.7 GB (2 parts) 57.3 GB
Too big Get file →

For 8 GB VRAM, use Q3_K_XL (13.15 GB file, ~15.8 GB memory at 8k context). For 12 GB VRAM, Q4_K_XL (17.56 GB file, ~20.2 GB memory). For 24 GB VRAM, Q6_K_XL (25.3 GB file, ~28 GB memory). BF16 requires 55.66 GB file and ~57.3 GB memory, split into two shards.

What GPU do you need for Tessera-27B-B-CJK?

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: the best quant that fits is Q3_K_XL (13.1 GB file, about 15.8 GB in use).
  • 24 GB VRAM: the best quant that fits is Q5_K_XL (20.9 GB file, about 23.6 GB in use).
  • 32 GB VRAM: the best quant that fits is Q6_K_XL (25.3 GB file, about 28.0 GB in use).
  • 48 GB VRAM: the best quant that fits is Q8_K_XL (31.5 GB file, about 34.1 GB in use).

Which Tessera-27B-B-CJK quant should you choose?

Q5_K_XL (20.9 GB) is the middle option in this repository. Closer to the original model than 4-bit, for a moderate increase in size. A good pick when you have memory to spare.

If memory is tight, the smallest file is Q3_K_XL at 13.1 GB, with the quality trade-offs that come with fewer bits. A reasonable choice when a 4-bit file will not fit, mostly on larger models. Quality loss is noticeable but often acceptable.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architectureqwen35
Parameters27.3B
Layers65
Trained context262,144 tokens
KV cache (fp16)260 KB per token

Source

Quantized byZelligeAI
FamilyTessera
Licenseapache-2.0
Downloads (HF)15,107
Last updated2026-10-14

Frequently asked questions about Tessera-27B-B-CJK

How much VRAM does Tessera-27B-B-CJK need?

About 23.6 GB at Q5_K_XL with an 8K context window. That is the 20.9 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 Tessera-27B-B-CJK on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 15.8 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Tessera-27B-B-CJK GGUF?

It depends on the quant: Q3_K_XL is 13.1 GB, Q4_K_XL is 17.6 GB, Q5_K_XL is 20.9 GB, Q6_K_XL is 25.3 GB, Q8_K_XL is 31.5 GB, BF16 is 54.7 GB.

What context length does Tessera-27B-B-CJK support?

The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 2.2 GB at 8K and 8.7 GB at 32K.

Where can I download Tessera-27B-B-CJK GGUF?

The files are on Hugging Face in the ZelligeAI/Tessera-27B-B-CJK-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 Tessera-27B-B-CJK 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 Tessera-27B-B-CJK on 8 GB VRAM?

No, the smallest quant requires ~15.8 GB memory at 8k context. 8 GB VRAM is insufficient.

Does the model support vision?

Yes, vision is detected, but image input has not been tested.

What license applies?

The model is released under the Apache‑2.0 license.

Other GGUFs of Tessera-27B-B-CJK

Similar models

Browse all →

More from ZelligeAI

All by ZelligeAI →

More Tessera models

All Tessera →

Other medium models

All medium models →