Nimbus-2B GGUF

Text-only llama.cpp artifacts for Nimbus-2B. Q5_K_M is the quality-first local default; Q4_K_M is listed only where it actually exists.

Choose a Nimbus model

Model	Best fit	Transformers	GGUF
Nimbus-2B	Fast drafting and focused edits	`Nimbus-Labs/Nimbus-2B`	`Nimbus-Labs/Nimbus-2B-GGUF`
Nimbus-4B	Balanced implementation and debugging	`Nimbus-Labs/Nimbus-4B`	`Nimbus-Labs/Nimbus-4B-GGUF`
Nimbus-9B v2.1	Deeper coding and reasoning	`Nimbus-Labs/Nimbus-9B-v2.1`	`Nimbus-Labs/Nimbus-9B-v2.1-GGUF`

The adjacent assets/nimbus-family-footprint.json contains the plotted values. Download size is not runtime memory: context cache and runtime buffers require additional capacity.

Downloads

File	Role	Bytes	SHA-256
`Nimbus-2B-BF16.gguf`	BF16 reference	3,775,708,544 (3.78 GB)	`4c35e21bec421799ba27fccb4ed560a534b243076c05fefb386b95d9e140123a`
`Nimbus-2B-Q4_K_M.gguf`	Memory-first	1,274,396,032 (1.27 GB)	`25aefd6d16c6af14c87028e3edc958069e8515eb179727e12ba5c4180c46f995`
`Nimbus-2B-Q5_K_M.gguf`	Quality-first default	1,411,120,512 (1.41 GB)	`25d813bfe0d655dfa3629322b332a3edd4d56c5d3b5f72a1fd245786b6770d03`

llama.cpp

Validated release runtime: llama.cpp b10007. Native thinking uses the supplied Qwen/Ornith chat template and DeepSeek-style reasoning parsing.

llama-server --model Nimbus-2B-Q5_K_M.gguf --ctx-size 65536 --n-gpu-layers all --reasoning-format deepseek

Evaluation

No public score is claimed in this card yet. Results will be added only after the exact released artifact, complete task set, and scorer outputs are bound to a release manifest.

Provenance and scope

Quantized from Nimbus-Labs/Nimbus-2B
Foundation: Qwen3.5
GGUF files are text-only; the Transformers repository retains the multimodal components.
Do not infer memory fit from file size alone; context KV cache and runtime buffers require additional memory.