CO

CohereLabs.command A Plus 05 2026 Bf16

CommandGGUF by DevQuasarUpdated 2026-06-07
Download on Hugging Face
Parameters218B
Context4883K
Layers32
Architecturecohere2-moe
Smallest quant79.7 GB
Quick answer

CohereLabs.command A Plus 05 2026 Bf16 needs about 133.8 GB of memory at BF16 (a 132.3 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 6 quants, from 79.7 GB (BF16) to 232.0 GB (BF16). Longer context needs more: at 32K tokens the KV cache alone adds about 4.3 GB.

Command A Plus 05 2026 Bf16 – 218B total parameters, 25B active, 128K context

Command A Plus 05 2026 Bf16 is a 218B total parameter model with 25B active parameters. It supports text and image inputs and is optimized for agentic, multilingual, and reasoning tasks. The model is released under Apache‑2.0.

Good at

  • Agentic reasoning
  • Multilingual text generation
  • Image‑text interaction

Limits

  • No vision support in GGUF version
  • Large VRAM requirement for full model
  • Requires llama.cpp cohere2‑moe support

Best for

  • Enterprise chatbots
  • Multilingual content creation
  • Complex reasoning workflows

Who it’s for. Developers building AI agents that need robust reasoning and multilingual capabilities. Suitable for teams with access to high‑memory GPUs and the ability to run llama.cpp with cohere2‑moe support.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
BF1679.7 GB (6 parts) 81.2 GB
Too big Get file →
BF16104.5 GB (7 parts) 106.1 GB
Too big Get file →
BF16132.3 GB (9 parts) 133.8 GB
Too big Get file →
BF16155.0 GB (11 parts) 156.5 GB
Too big Get file →
BF16179.1 GB (12 parts) 180.7 GB
Too big Get file →
BF16232.0 GB (16 parts) 233.5 GB
Too big Get file →

The BF16 quantizations range from 79.66 GB to 231.96 GB. For 8 GB VRAM use the 79.66 GB shard set (6 shards). For 12 GB VRAM use the 104.54 GB set (7 shards). For 16 GB VRAM use the 132.27 GB set (9 shards). Estimates are approximate.

What GPU do you need for CohereLabs.command A Plus 05 2026 Bf16?

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: no quant fits at an 8K context.

Which CohereLabs.command A Plus 05 2026 Bf16 quant should you choose?

BF16 (132.3 GB) is the middle option in this repository. The unquantized model. It gives original quality and needs the most memory, so it is mainly for conversion or reference.

If memory is tight, the smallest file is BF16 at 79.7 GB, with the quality trade-offs that come with fewer bits. The unquantized model. It gives original quality and needs the most memory, so it is mainly for conversion or reference.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturecohere2-moe
Parameters218B (approximate, mixture of experts)
Layers32
Trained context5,000,000 tokens
KV cache (fp16)128 KB per token

Source

Quantized byDevQuasar
FamilyCommand
Licenseapache-2.0
Downloads (HF)762
Last updated2026-06-07

Frequently asked questions about CohereLabs.command A Plus 05 2026 Bf16

How much VRAM does CohereLabs.command A Plus 05 2026 Bf16 need?

About 133.8 GB at BF16 with an 8K context window. That is the 132.3 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 CohereLabs.command A Plus 05 2026 Bf16 on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 81.2 GB, so you would need a larger GPU or CPU offloading.

What is the file size of CohereLabs.command A Plus 05 2026 Bf16 GGUF?

It depends on the quant: BF16 is 79.7 GB, BF16 is 104.5 GB, BF16 is 132.3 GB, BF16 is 155.0 GB, BF16 is 179.1 GB, BF16 is 232.0 GB.

What context length does CohereLabs.command A Plus 05 2026 Bf16 support?

The GGUF header lists a trained context length of 5,000,000 tokens (4883K). 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 CohereLabs.command A Plus 05 2026 Bf16 GGUF?

The files are on Hugging Face in the DevQuasar/CohereLabs.command-a-plus-05-2026-bf16-GGUF-EXPERIMENTAL repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does CohereLabs.command A Plus 05 2026 Bf16 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 this model on 8 GB VRAM?

Use the 79.66 GB BF16 quantization split into 6 shards. It will require about 81.2 GB memory at 8k context, so it is not feasible on 8 GB VRAM.

Does the GGUF version support image inputs?

The GGUF quantized files do not include vision support; only text is supported in this build.

What license applies to this model?

The model is licensed under Apache‑2.0.

Other GGUFs of command-a-plus-05-2026-bf16

Similar models

Browse all →

More from DevQuasar

All by DevQuasar →

More Command models

All Command →

Other large models

All large models →