---

base_model:

• poolside/Laguna-S-2.1

base_model_relation: quantized

---

<p align="center">

<img alt="poolside-banner" src="https://poolside.ai/assets/laguna/laguna-s-2-1-banner.svg" width="800px">

</p>

<p align="center">

<a href="https://openrouter.ai/poolside/laguna-s-2.1"><strong>Use on OpenRouter</strong></a> ·

<a href="https://vercel.com/ai-gateway/models/laguna-s-2.1"><strong>Use on Vercel AI Gateway</strong></a> ·

<a href="https://poolside.ai/blog/introducing-laguna-s-2-1"><strong>Release blog post</strong></a>

</p>

<br>

Laguna S 2.1 GGUF

GGUF conversions of Laguna S 2.1 for llama.cpp, plus the DFlash speculative-decoding draft model. See the base model card for architecture details, license, and usage guidance.

Files

| File | Size | Notes |

|------|------|-------|

| laguna-s-2.1-F16.gguf | 235 GB | full precision |

| laguna-s-2.1-Q8_0.gguf | 129 GB | routed experts Q8_0, signal path (attention, shared experts, embeddings) kept BF16 |

| laguna-s-2.1-Q4_K_M.gguf | 68 GB | routed experts Q4_K (imatrix), signal path kept Q8_0 |

| laguna-s-2.1-DFlash-BF16.gguf | 2.2 GB | DFlash drafter for speculative decoding |

| laguna-s-2.1.imatrix | 0.4 GB | importance matrix used for the K-quants |

Serving

Serve with Poolside's llama.cpp fork, branch

laguna, which carries full

Laguna support including DFlash speculative decoding. (Base Laguna support is also

in upstream review: ggml-org/llama.cpp#25165.)

git clone --branch laguna https://github.com/poolsideai/llama.cpp
cd llama.cpp && cmake -B build && cmake --build build -j

./build/bin/llama-server -m laguna-s-2.1-Q4_K_M.gguf --jinja --port 8000

# with DFlash speculative decoding:
./build/bin/llama-server -m laguna-s-2.1-Q4_K_M.gguf \
  -md laguna-s-2.1-DFlash-BF16.gguf \
  --spec-type draft-dflash --spec-draft-n-max 15 -fa on --jinja --port 8000

Context length

These GGUFs ship configured for a 262,144-token (256K) context window. This is the configuration we recommend for best output quality.

The weights are native 1M checkpoints: training included a long-context extension stage up to 1,048,576 tokens. To use more than 256K of context with llama.cpp, override the rope configuration at load time:

--ctx-size 1048576 --rope-scaling yarn --rope-scale 128 --yarn-orig-ctx 8192

You may experience quality degradation with the 1M configuration. If you use it, we recommend sampling with --temp 0.7 --top-p 0.95.

This release also corrects the embedded yarn_attn_factor metadata (now 1.0; llama.cpp derives the YaRN attention scaling internally).