Laguna S 2.1 initial release

Co-authored-by: varunrandery <varunrandery@users.noreply.huggingface.co>
Co-authored-by: Jiminator <Jiminator@users.noreply.huggingface.co>

.eval_results/deepswe.yaml

@@ -0,0 +1,7 @@
+- dataset:
+    id: datacurve/deep-swe
+    task_id: deepswe_%_resolved
+  value: 40.4
+  source:
+    url: https://huggingface.co/poolside/Laguna-S-2.1
+    name: Model Card

.eval_results/swe-bench_multilingual.yaml

@@ -0,0 +1,7 @@
+- dataset:
+    id: SWE-bench/SWE-bench_Multilingual
+    task_id: swe_bench_%_resolved
+  value: 78.5
+  source:
+    url: https://huggingface.co/poolside/Laguna-S-2.1
+    name: Model Card

.eval_results/swe-bench_pro.yaml

@@ -0,0 +1,7 @@
+- dataset:
+    id: ScaleAI/SWE-bench_Pro
+    task_id: SWE_Bench_Pro
+  value: 59.4
+  source:
+    url: https://huggingface.co/poolside/Laguna-S-2.1
+    name: Model Card

.gitattributes

@@ -0,0 +1,35 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text

LICENSE.md

@@ -0,0 +1,49 @@
+OpenMDW License Agreement, version 1.1 (OpenMDW-1.1)
+
+By exercising rights granted to you under this agreement, you accept and agree
+to its terms.
+
+As used in this agreement, "Model Materials" means the materials provided to
+you under this agreement, consisting of: (1) one or more machine learning
+models (including architecture and parameters); and (2) all related artifacts
+(including associated data, documentation and software) that are provided to
+you hereunder.
+
+Subject to your compliance with this agreement, permission is hereby granted,
+free of charge, to deal in the Model Materials without restriction, including
+under all copyright, patent, database, and trade secret rights included or
+embodied therein.
+
+If you distribute any portion of the Model Materials, you shall retain in your
+distribution (1) a copy of this agreement, and (2) all copyright notices and
+other notices of origin included in the Model Materials that are applicable to
+your distribution.
+
+If you file, maintain, or voluntarily participate in a lawsuit against any
+person or entity asserting that the Model Materials directly or indirectly
+infringe any patent or copyright, then all rights and grants made to you
+hereunder are terminated, unless that lawsuit was in response to a
+corresponding lawsuit first brought against you.
+
+This agreement does not impose any restrictions or obligations with respect to
+any use, modification, or sharing of any outputs generated by using the Model
+Materials.
+
+THE MODEL MATERIALS ARE PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
+OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE, TITLE, NONINFRINGEMENT, ACCURACY, OR THE
+ABSENCE OF LATENT OR OTHER DEFECTS OR ERRORS, WHETHER OR NOT DISCOVERABLE, ALL
+TO THE GREATEST EXTENT PERMISSIBLE UNDER APPLICABLE LAW.
+
+YOU ARE SOLELY RESPONSIBLE FOR (1) CLEARING RIGHTS OF OTHER PERSONS THAT MAY
+APPLY TO THE MODEL MATERIALS OR ANY USE THEREOF, INCLUDING WITHOUT LIMITATION
+ANY PERSON'S COPYRIGHTS OR OTHER RIGHTS INCLUDED OR EMBODIED IN THE MODEL
+MATERIALS; (2) OBTAINING ANY NECESSARY CONSENTS, PERMISSIONS OR OTHER RIGHTS
+REQUIRED FOR ANY USE OF THE MODEL MATERIALS; OR (3) PERFORMING ANY DUE
+DILIGENCE OR UNDERTAKING ANY OTHER INVESTIGATIONS INTO THE MODEL MATERIALS OR
+ANYTHING INCORPORATED OR EMBODIED THEREIN.
+
+IN NO EVENT SHALL THE PROVIDERS OF THE MODEL MATERIALS BE LIABLE FOR ANY CLAIM,
+DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
+OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE MODEL MATERIALS, THE
+USE THEREOF OR OTHER DEALINGS THEREIN.

README.md

@@ -0,0 +1,182 @@
+---
+library_name: transformers
+inference: false
+extra_gated_description: >-
+  To learn more about how we process your personal data, please read our <a
+  href="https://poolside.ai/legal/privacy">Privacy Policy</a>.
+tags:
+- laguna-s-2.1
+- vllm
+license: openmdw-1.1
+pipeline_tag: text-generation
+---
+
+<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
+
+Laguna S 2.1 is a 118B total parameter Mixture-of-Experts model with 8B activated
+parameters per token, designed for agentic coding and long-horizon work. It sits
+between [Laguna XS 2.1](https://huggingface.co/poolside/Laguna-XS-2.1) (33B-A3B) and
+Laguna M.1 (225B-A23B) in the Laguna series and shares the family recipe: a
+token-choice router with softplus gating over 256 routed experts plus one shared
+expert, grouped-query attention, and interleaved full/sliding-window attention.
+
+## Highlights
+
+- **Mixed SWA and global attention layout**: 48 layers in a 1:3 global-to-SWA ratio
+  (12 global attention layers, 36 sliding-window layers, window 512), with softplus
+  attention gating and per-layer-type rotary scales
+- **1M context**: 1,048,576-token context window
+- **Native reasoning support**: interleaved thinking between tool calls, with
+  per-request control via `enable_thinking`
+- **Speculative decoding**: a trained
+  [DFlash draft model](https://huggingface.co/poolside/Laguna-S-2.1-DFlash) is available
+  for lower-latency serving
+- **Quantized variants**:
+  [FP8](https://huggingface.co/poolside/Laguna-S-2.1-FP8),
+  [NVFP4](https://huggingface.co/poolside/Laguna-S-2.1-NVFP4),
+  [INT4](https://huggingface.co/poolside/Laguna-S-2.1-INT4) and
+  [GGUF](https://huggingface.co/poolside/Laguna-S-2.1-GGUF)
+- **OpenMDW-1.1 license**: Use and modify the model and associated materials freely
+  for commercial and non-commercial purposes
+  ([learn more about OpenMDW](https://openmdw.ai/))
+
+## Model overview
+
+- Number of parameters: 118B total, ~8B activated per token
+- Layers: 48 (12 global attention, 36 sliding-window attention)
+- Experts: 256 routed (top-10) plus 1 shared expert
+- Attention: grouped-query, 8 KV heads, head dim 128; per-head softplus output gating
+- Sliding window: 512 tokens
+- Context window: 1,048,576 tokens
+- Vocabulary: 100,352 tokens (Laguna family tokenizer)
+- Modality: text-to-text
+- Reasoning: interleaved thinking with preserved thinking
+
+## Benchmark results
+
+<p align="center">
+  <img alt="benchmarks" src="https://poolside.ai/assets/laguna/laguna-s-2-1-chart.svg" width="800px">
+</p>
+
+| Model | Size | Terminal-Bench 2.1 | SWE-bench Multilingual | SWE-Bench Pro (Public Dataset) | DeepSWE | SWE Atlas (Codebase QnA) | Toolathlon Verified |
+|---|---|---|---|---|---|---|---|
+| **Laguna S 2.1** | 118B-A8B | **70.2%** | **78.5%** | **59.4%** | **40.4%** | **46.2%** | **49.7%** |
+| Tencent Hy3 | 295B-A21B | 71.7% | 75.8% | 57.9% | - | - | - |
+| Inkling | 975B-A41B | 63.8% | - | 54.3% | - | - | 45.5%* |
+| Nemotron 3 Ultra | 550B-A55B | 56.4% | 67.7% | - | - | - | 34.3%* |
+| DeepSeek-V4-Pro Max | 1.6T-A49B | 64.0%* | 76.2% | 55.4% | 9.0%* | 27.2%* | 55.9%* |
+| Kimi K3 | 2800B-A50B | 88.3% | - | - | 69% | - | - |
+| Qwen 3.7 Max | - | 74.5%* | 78.3% | 60.6% | - | - | - |
+| Muse Spark 1.1 | - | 80% | - | 61.5% | 53.3% | 42.2%* | 75.6% |
+| Claude Fable 5 | - | 88% | - | 80.3% | 70% | - | - |
+
+Benchmarks as of 21 July 2026. Laguna S 2.1 in **bold**; a dash (-) marks a benchmark a model was not evaluated on. Scores marked * are as reported by third parties: Terminal-Bench 2.1 and DeepSWE via Artificial Analysis, SWE Atlas via Scale AI's official leaderboard, and Toolathlon Verified via its official leaderboard. Full evaluation trajectories: [trajectories.poolside.ai](https://trajectories.poolside.ai).
+
+## Usage
+
+Laguna S 2.1 uses the same `laguna` architecture as Laguna XS 2.1, so the same
+engine integrations apply (vLLM, SGLang, Transformers, TRT-LLM, llama.cpp). At 118B
+parameters the BF16 checkpoint needs multiple GPUs (roughly 236GB of weights);
+quantized variants reduce this substantially.
+
+### vLLM
+
+```shell
+vllm serve \
+    --model poolside/Laguna-S-2.1 \
+    --tensor-parallel-size 4 \
+    --tool-call-parser poolside_v1 \
+    --reasoning-parser poolside_v1 \
+    --enable-auto-tool-choice \
+    --served-model-name laguna \
+    --default-chat-template-kwargs '{"enable_thinking": true}'
+```
+
+> [!NOTE]
+> **Optional: speculative decoding with DFlash.** Pair with the
+> [Laguna S 2.1 DFlash draft model](https://huggingface.co/poolside/Laguna-S-2.1-DFlash)
+> by adding
+> `--speculative-config '{"model":"poolside/Laguna-S-2.1-DFlash","num_speculative_tokens":7,"method":"dflash"}'`.
+
+### SGLang
+
+```shell
+python -m sglang.launch_server \
+  --model-path poolside/Laguna-S-2.1 \
+  --tp-size 4 \
+  --reasoning-parser poolside_v1 \
+  --tool-call-parser poolside_v1 \
+  --trust-remote-code
+```
+
+### TRT-LLM
+
+```shell
+trtllm-serve poolside/Laguna-S-2.1 --trust-remote-code \
+    --tool_parser poolside_v1 --reasoning_parser laguna
+```
+
+Note the flag names differ from vLLM's (`--tool_parser`, and the reasoning parser
+is `laguna`, not `poolside_v1`).
+
+### llama.cpp
+
+GGUF conversions are available at
+[poolside/Laguna-S-2.1-GGUF](https://huggingface.co/poolside/Laguna-S-2.1-GGUF).
+Serve with poolside's llama.cpp fork, branch
+[`laguna`](https://github.com/poolsideai/llama.cpp/tree/laguna), which carries
+full Laguna support including DFlash speculative decoding. (Base Laguna support
+is also in upstream review:
+[ggml-org/llama.cpp#25165](https://github.com/ggml-org/llama.cpp/pull/25165).)
+
+```shell
+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
+```
+
+## Controlling reasoning
+
+Laguna S 2.1 has native reasoning support and works best with *preserved thinking*:
+keep `reasoning_content` from prior assistant messages in the message history.
+The model will generally reason before calling tools and between tool calls, and
+may stop reasoning in follow-up steps if prior thinking blocks are dropped.
+
+Thinking is controlled per request via the chat template:
+
+```python
+extra_body={"chat_template_kwargs": {"enable_thinking": False}}
+```
+
+or at the server level with
+`--default-chat-template-kwargs '{"enable_thinking": true}'`. For agentic coding
+use cases we recommend enabling thinking and preserving reasoning in the message
+history.
+
+## License
+
+This model is licensed under the [OpenMDW-1.1 License](https://huggingface.co/poolside/Laguna-S-2.1/blob/main/LICENSE.md).
+
+## Intended and Responsible Use
+
+Laguna S 2.1 is designed for software engineering and agentic coding use cases, and you are responsible for confirming that it is appropriate for your intended application. Laguna S 2.1 is subject to the [OpenMDW-1.1 License](https://huggingface.co/poolside/Laguna-S-2.1/blob/main/LICENSE.md), and should be used consistently with Poolside's [Acceptable Use Policy](https://poolside.ai/legal/acceptable-use-policy). We advise against circumventing Laguna S 2.1 safety guardrails without implementing substantially equivalent mitigations appropriate for your use case.
+
+Please report security vulnerabilities or safety concerns to [security@poolside.ai](mailto:security@poolside.ai).

chat_template.jinja

@@ -0,0 +1,93 @@
+{#- Iteration on laguna_glm_thinking_v8/chat_template.jinja -#}
+{#- No formatting instructions -#}
+{{- "〈|EOS|〉" -}}
+{%- set enable_thinking = enable_thinking | default(false) -%}
+{%- set add_generation_prompt = add_generation_prompt | default(false) -%}
+
+{#- ───── header (system message) ───── -#}
+{#- A caller-supplied system message with empty content opts out of the default below, producing no <system> block — used to train without a system message. -#}
+{%- set system_message = "You are a helpful, conversationally-fluent assistant made by Poolside. You are here to be helpful to users through natural language conversations." -%}
+{%- if messages and messages[0].role == "system" -%}
+  {%- set system_message = messages[0].content -%}
+  {%- set messages = messages[1:] -%}
+{%- endif -%}
+
+{%- set has_sys = system_message and system_message.strip() -%}
+{%- if has_sys or tools or enable_thinking -%}
+  {{- "<system>" -}}
+
+  {%- if has_sys -%}
+    {{- system_message.rstrip() -}}
+    {%- if tools -%}{{- "\n\n" -}}{%- endif -%}
+  {%- endif -%}
+
+  {%- if tools -%}
+    {{- "### Tools\n\n" -}}
+    {{- "You may call functions to assist with the user query.\n" -}}
+    {{- "All available function signatures are listed below:\n" -}}
+    {{- "<available_tools>\n" -}}
+    {%- for tool in tools -%}
+      {{- (tool | tojson) ~ "\n" -}}
+    {%- endfor -%}
+    {{- "</available_tools>" -}}
+  {%- endif -%}
+
+  {{- "</system>\n" -}}
+{%- endif -%}
+
+{#- ───── main loop ───── -#}
+{%- for message in messages -%}
+  {%- set content = message.content if message.content is string else "" -%}
+  {%- if message.role == "user" -%}
+    {{- "<user>" + content + "</user>\n" -}}
+  {%- elif message.role == "assistant" -%}
+    {%- generation -%}
+      {{- "<assistant>" -}}
+      {#- Extract reasoning content from message.reasoning (vLLM field name) or message.reasoning_content -#}
+      {%- set reasoning_content = '' -%}
+      {%- if message.reasoning is string -%}
+        {%- set reasoning_content = message.reasoning -%}
+      {%- elif message.reasoning_content is string -%}
+        {%- set reasoning_content = message.reasoning_content -%}
+      {%- endif -%}
+      {#- Display reasoning content for all messages if enable_thinking -#}
+      {%- if enable_thinking -%}
+        {{- '<think>' + reasoning_content + '</think>' -}}
+      {%- else -%}
+        {{- '</think>' -}}
+      {%- endif -%}
+      {#- Display main content (trailing newline only when no tool_calls follow) -#}
+      {%- if content -%}
+        {{- content -}}
+      {%- endif -%}
+      {%- if message.tool_calls -%}
+        {%- for tool_call in message.tool_calls -%}
+          {%- set function_data = tool_call.function -%}
+          {{- '<tool_call>' + function_data.name -}}
+          {%- set _args = function_data.arguments -%}
+          {%- for k, v in _args.items() -%}
+            {{- "<arg_key>" ~ k ~ "</arg_key>" -}}
+            {{- "<arg_value>" -}}{{- v | tojson(ensure_ascii=False) if v is not string else v -}}{{- "</arg_value>" -}}
+          {%- endfor -%}
+          {{- "</tool_call>" -}}
+        {%- endfor -%}
+      {%- endif -%}
+      {{- "</assistant>\n" -}}
+    {%- endgeneration -%}
+  {%- elif message.role == "tool" -%}
+    {{- "<tool_response>" + content + "</tool_response>\n" -}}
+  {%- elif message.role == "system" -%}
+    {#- Render additional system messages (the first one, if any, is handled separately in the header and was sliced off above) -#}
+    {{- "<system>" + content + "</system>\n" -}}
+  {%- endif -%}
+{%- endfor -%}
+{#- ───── generation prompt ───── -#}
+{%- if add_generation_prompt -%}
+  {{- "<assistant>" -}}
+  {#- ───── Include reasoning mode directive ───── -#}
+  {%- if enable_thinking -%}
+    {{- '<think>' -}}
+  {%- else -%}
+    {{- '</think>' -}}
+  {%- endif -%}
+{%- endif -%}

config.json

@@ -0,0 +1,262 @@
+{
+  "architectures": [
+    "LagunaForCausalLM"
+  ],
+  "auto_map": {
+    "AutoConfig": "configuration_laguna.LagunaConfig",
+    "AutoModelForCausalLM": "modeling_laguna.LagunaForCausalLM"
+  },
+  "model_type": "laguna",
+  "vocab_size": 100352,
+  "hidden_size": 3072,
+  "intermediate_size": 12288,
+  "num_hidden_layers": 48,
+  "num_attention_heads": 48,
+  "num_key_value_heads": 8,
+  "head_dim": 128,
+  "max_position_embeddings": 1048576,
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "rms_norm_eps": 1e-06,
+  "num_experts": 256,
+  "num_experts_per_tok": 10,
+  "moe_intermediate_size": 1024,
+  "shared_expert_intermediate_size": 1024,
+  "norm_topk_prob": true,
+  "router_aux_loss_coef": 0.0,
+  "decoder_sparse_step": 1,
+  "mlp_only_layers": [
+    0
+  ],
+  "bos_token_id": 2,
+  "eos_token_id": [
+    2,
+    24
+  ],
+  "pad_token_id": 9,
+  "tie_word_embeddings": false,
+  "use_cache": true,
+  "torch_dtype": "bfloat16",
+  "gating": "per-head",
+  "sliding_window": 512,
+  "rope_parameters": {
+    "full_attention": {
+      "rope_theta": 500000.0,
+      "rope_type": "yarn",
+      "factor": 128.0,
+      "original_max_position_embeddings": 8192,
+      "beta_slow": 1.0,
+      "beta_fast": 32.0,
+      "attention_factor": 1.4852030263919618,
+      "partial_rotary_factor": 0.5
+    },
+    "sliding_attention": {
+      "rope_type": "default",
+      "rope_theta": 10000.0,
+      "partial_rotary_factor": 1.0
+    }
+  },
+  "layer_types": [
+    "full_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention"
+  ],
+  "moe_apply_router_weight_on_input": false,
+  "mlp_layer_types": [
+    "dense",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse"
+  ],
+  "gating_types": [
+    "per_head",
+    "per_head",
+    "per_head",
+    "per_head",
+    "per_head",
+    "per_head",
+    "per_head",
+    "per_head",
+    "per_head",
+    "per_head",
+    "per_head",
+    "per_head",
+    "per_head",
+    "per_head",
+    "per_head",
+    "per_head",
+    "per_head",
+    "per_head",
+    "per_head",
+    "per_head",
+    "per_head",
+    "per_head",
+    "per_head",
+    "per_head",
+    "per_head",
+    "per_head",
+    "per_head",
+    "per_head",
+    "per_head",
+    "per_head",
+    "per_head",
+    "per_head",
+    "per_head",
+    "per_head",
+    "per_head",
+    "per_head",
+    "per_head",
+    "per_head",
+    "per_head",
+    "per_head",
+    "per_head",
+    "per_head",
+    "per_head",
+    "per_head",
+    "per_head",
+    "per_head",
+    "per_head",
+    "per_head"
+  ],
+  "moe_routed_scaling_factor": 2.5,
+  "num_attention_heads_per_layer": [
+    48,
+    72,
+    72,
+    72,
+    48,
+    72,
+    72,
+    72,
+    48,
+    72,
+    72,
+    72,
+    48,
+    72,
+    72,
+    72,
+    48,
+    72,
+    72,
+    72,
+    48,
+    72,
+    72,
+    72,
+    48,
+    72,
+    72,
+    72,
+    48,
+    72,
+    72,
+    72,
+    48,
+    72,
+    72,
+    72,
+    48,
+    72,
+    72,
+    72,
+    48,
+    72,
+    72,
+    72,
+    48,
+    72,
+    72,
+    72
+  ],
+  "moe_router_logit_softcapping": 0.0
+}

configuration_laguna.py

@@ -0,0 +1,249 @@
+# ruff: noqa
+# Copyright 2025 Poolside and the HuggingFace Inc. team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+from transformers.configuration_utils import PreTrainedConfig
+from transformers.modeling_rope_utils import RopeParameters
+from transformers.utils.import_utils import is_causal_conv1d_available, is_flash_linear_attention_available
+
+
+class LagunaConfig(PreTrainedConfig):
+    r"""
+    Configuration class for Laguna model.
+
+    Laguna is Poolside's MoE architecture with:
+    - Attention output gating (softplus gate)
+    - Sigmoid routing instead of softmax
+    - No QKV bias
+    - Explicit head_dim parameter
+
+    Args:
+        head_dim (`int`, *optional*, defaults to 128):
+            Dimension of attention heads. Laguna uses explicit head_dim rather than
+            computing it from hidden_size // num_attention_heads.
+        qkv_bias (`bool`, *optional*, defaults to `False`):
+            Whether to add bias to QKV projections. Laguna uses no QKV bias.
+        attention_bias (`bool`, *optional*, defaults to `False`):
+            Whether to add bias to attention output projection. Laguna uses no attention bias.
+        gating (`bool` or `str`, *optional*, defaults to `True`):
+            Attention output gating mode. When ``True`` or ``"per-element"`` a g_proj
+            linear layer with output size ``num_attention_heads * head_dim`` is added
+            and ``attn_output = attn_output * softplus(g_proj(x))``. When ``"per-head"``
+            g_proj has output size ``num_attention_heads`` and the gate broadcasts across
+            ``head_dim``. When ``False`` no gating is applied.
+        partial_rotary_factor (`float`, *optional*):
+            Fraction of head_dim to apply rotary embeddings to. When set, this value is
+            injected into ``rope_parameters`` (and ``swa_rope_parameters``) if not already
+            specified there. When ``None`` the default behaviour of the rope implementation
+            is used (typically full rotary).
+        num_attention_heads_per_layer (`list[int]`, *optional*):
+            Optional per-layer override for ``num_attention_heads``. When provided the list
+            length must equal ``num_hidden_layers`` and each entry is the head count used by
+            that layer. When ``None`` every layer uses ``num_attention_heads``.
+        vocab_size (`int`, *optional*, defaults to 100352):
+            Vocabulary size of the Laguna model.
+        hidden_size (`int`, *optional*, defaults to 2048):
+            Dimension of the hidden representations.
+        intermediate_size (`int`, *optional*, defaults to 8192):
+            Dimension of the MLP representations for dense layers.
+        num_hidden_layers (`int`, *optional*, defaults to 48):
+            Number of hidden layers in the Transformer.
+        num_attention_heads (`int`, *optional*, defaults to 32):
+            Number of attention heads.
+        num_key_value_heads (`int`, *optional*, defaults to 8):
+            Number of key-value heads for GQA.
+        max_position_embeddings (`int`, *optional*, defaults to 4096):
+            Maximum sequence length.
+        rms_norm_eps (`float`, *optional*, defaults to 1e-6):
+            Epsilon for RMSNorm layers.
+        sliding_window (`int`, *optional*):
+            Sliding window attention size. Used by layers whose type in ``layer_types``
+            is ``"sliding_attention"``. When ``None``, all layers use full attention.
+        layer_types (`list[str]`, *optional*):
+            Per-layer attention type. Each element should be ``"sliding_attention"`` or
+            ``"full_attention"``. Length must equal ``num_hidden_layers``. When ``None``,
+            all layers default to global attention.
+        swa_attention_sink_enabled (`bool`, *optional*, defaults to `False`):
+            Whether to enable learnable attention sinks on sliding-window attention layers.
+            When enabled, a per-head bias parameter is added that allows the model to attend
+            to position 0 even when it falls outside the sliding window.
+        swa_rope_parameters (`RopeParameters`, *optional*):
+            Separate RoPE configuration for sliding-window attention layers. When ``None``,
+            SWA layers use the same RoPE as global attention layers.
+        num_experts (`int`, *optional*, defaults to 256):
+            Number of routed experts.
+        num_experts_per_tok (`int`, *optional*, defaults to 16):
+            Number of experts selected per token (top-k).
+        moe_intermediate_size (`int`, *optional*, defaults to 1024):
+            Intermediate size of routed experts.
+        shared_expert_intermediate_size (`int`, *optional*, defaults to 1024):
+            Intermediate size of the shared expert.
+        norm_topk_prob (`bool`, *optional*, defaults to `True`):
+            Whether to normalize top-k routing probabilities.
+        decoder_sparse_step (`int`, *optional*, defaults to 1):
+            Frequency of MoE layers (1 = every layer is MoE after mlp_only_layers).
+        mlp_only_layers (`list[int]`, *optional*, defaults to `[0]`):
+            Layer indices that use dense MLP instead of MoE.
+        router_aux_loss_coef (`float`, *optional*, defaults to 0.001):
+            Auxiliary loss coefficient for load balancing.
+        moe_routed_scaling_factor (`float`, *optional*, defaults to 1.0):
+            Scalar multiplier applied to the routed-expert output before combining with the
+            shared-expert output.
+        moe_apply_router_weight_on_input (`bool`, *optional*, defaults to `False`):
+            When ``True`` the top-k routing weights are multiplied into each expert's input
+            rather than its output. Matches the numerical form used by the trained checkpoint.
+        moe_router_logit_softcapping (`float`, *optional*, defaults to 0.0):
+            Optional soft-capping value ``c`` applied to router logits as
+            ``x = tanh(x / c) * c`` before sigmoid + top-k. Disabled when ``0``.
+        rope_parameters (`RopeParameters`, *optional*):
+            RoPE configuration. Defaults to rope_theta=500000.0.
+    """
+
+    model_type = "laguna"
+    keys_to_ignore_at_inference = ["past_key_values"]
+    # PreTrainedConfig in transformers v5 no longer auto-declares these; subclasses
+    # opt in by providing class-level annotations with defaults.
+    pad_token_id: int | None = None
+    bos_token_id: int | None = None
+    eos_token_id: int | list[int] | None = None
+    base_model_tp_plan = {
+        "layers.*.self_attn.q_proj": "colwise",
+        "layers.*.self_attn.k_proj": "colwise",
+        "layers.*.self_attn.v_proj": "colwise",
+        "layers.*.self_attn.g_proj": "colwise",  # Laguna-specific gating projection
+        "layers.*.self_attn.o_proj": "rowwise",
+        "layers.*.mlp.gate_proj": "colwise",
+        "layers.*.mlp.up_proj": "colwise",
+        "layers.*.mlp.down_proj": "rowwise",
+    }
+    base_model_pp_plan = {
+        "embed_tokens": (["input_ids"], ["inputs_embeds"]),
+        "layers": (["hidden_states", "attention_mask"], ["hidden_states"]),
+        "norm": (["hidden_states"], ["hidden_states"]),
+    }
+
+    def __init__(
+        self,
+        vocab_size: int = 100352,
+        hidden_size: int = 2048,
+        intermediate_size: int = 8192,
+        num_hidden_layers: int = 48,
+        num_attention_heads: int = 32,
+        num_key_value_heads: int = 8,
+        head_dim: int = 128,
+        qkv_bias: bool = False,
+        attention_bias: bool = False,
+        gating: bool | str = True,
+        hidden_act: str = "silu",
+        max_position_embeddings: int = 4096,
+        initializer_range: float = 0.02,
+        rms_norm_eps: float = 1e-6,
+        use_cache: bool = True,
+        tie_word_embeddings: bool = False,
+        rope_parameters: RopeParameters | dict[str, RopeParameters] | None = None,
+        partial_rotary_factor: float | None = None,
+        attention_dropout: float = 0.0,
+        sliding_window: int | None = None,
+        layer_types: list[str] | None = None,
+        num_attention_heads_per_layer: list[int] | None = None,
+        swa_attention_sink_enabled: bool = False,
+        swa_rope_parameters: RopeParameters | None = None,
+        num_experts: int = 256,
+        num_experts_per_tok: int = 16,
+        moe_intermediate_size: int = 1024,
+        shared_expert_intermediate_size: int = 1024,
+        norm_topk_prob: bool = True,
+        decoder_sparse_step: int = 1,
+        mlp_only_layers: list[int] | None = None,
+        router_aux_loss_coef: float = 0.001,
+        moe_routed_scaling_factor: float = 1.0,
+        moe_apply_router_weight_on_input: bool = False,
+        moe_router_logit_softcapping: float = 0.0,
+        output_router_logits: bool = False,
+        **kwargs,
+    ):
+        # Default mlp_only_layers: first layer is dense (moe_first_k_dense_replace=1)
+        if mlp_only_layers is None:
+            mlp_only_layers = [0]
+
+        # Default layer_types: all layers use full attention (Laguna-M). Laguna-XS
+        # ships an explicit list with a mix of "full_attention" and "sliding_attention".
+        # Downstream mask builders (``create_masks_for_generate``) iterate
+        # ``layer_types``, so it must be a list — not left as ``None``.
+        if layer_types is None:
+            layer_types = ["full_attention"] * num_hidden_layers
+
+        # Default rope_parameters with Laguna's theta
+        if rope_parameters is None:
+            rope_parameters = {"rope_type": "default", "rope_theta": 500000.0}
+
+        # config.json stores SWA rope nested in rope_parameters["sliding_attention"]
+        # and carries no top-level swa_rope_parameters. Derive it here, else the
+        # sliding-window layers silently reuse the full-attention rope.
+        if swa_rope_parameters is None and isinstance(rope_parameters, dict):
+            swa_rope_parameters = rope_parameters.get("sliding_attention")
+
+        # If ``partial_rotary_factor`` is set at the top level, inject it into any
+        # rope dict that does not already carry one so the rotary embedding picks
+        # it up consistently for both full-attention and SWA layers.
+        if partial_rotary_factor is not None:
+            if isinstance(rope_parameters, dict) and "partial_rotary_factor" not in rope_parameters:
+                rope_parameters = {**rope_parameters, "partial_rotary_factor": partial_rotary_factor}
+            if isinstance(swa_rope_parameters, dict) and "partial_rotary_factor" not in swa_rope_parameters:
+                swa_rope_parameters = {
+                    **swa_rope_parameters,
+                    "partial_rotary_factor": partial_rotary_factor,
+                }
+
+        self.vocab_size = vocab_size
+        self.hidden_size = hidden_size
+        self.intermediate_size = intermediate_size
+        self.num_hidden_layers = num_hidden_layers
+        self.num_attention_heads = num_attention_heads
+        self.num_key_value_heads = num_key_value_heads
+        self.head_dim = head_dim
+        self.qkv_bias = qkv_bias
+        self.attention_bias = attention_bias
+        self.gating = gating
+        self.hidden_act = hidden_act
+        self.max_position_embeddings = max_position_embeddings
+        self.initializer_range = initializer_range
+        self.rms_norm_eps = rms_norm_eps
+        self.use_cache = use_cache
+        self.rope_parameters = rope_parameters
+        self.partial_rotary_factor = partial_rotary_factor
+        self.attention_dropout = attention_dropout
+        # Sliding window attention arguments
+        self.sliding_window = sliding_window
+        self.layer_types = layer_types
+        self.num_attention_heads_per_layer = num_attention_heads_per_layer
+        self.swa_attention_sink_enabled = swa_attention_sink_enabled
+        self.swa_rope_parameters = swa_rope_parameters
+        # MoE arguments
+        self.num_experts = num_experts
+        self.num_experts_per_tok = num_experts_per_tok
+        self.moe_intermediate_size = moe_intermediate_size
+        self.shared_expert_intermediate_size = shared_expert_intermediate_size
+        self.norm_topk_prob = norm_topk_prob
+        self.decoder_sparse_step = decoder_sparse_step
+        self.mlp_only_layers = mlp_only_layers
+        self.router_aux_loss_coef = router_aux_loss_coef
+        self.moe_routed_scaling_factor = moe_routed_scaling_factor
+        self.moe_apply_router_weight_on_input = moe_apply_router_weight_on_input
+        self.moe_router_logit_softcapping = moe_router_logit_softcapping
+        self.output_router_logits = output_router_logits
+
+        super().__init__(tie_word_embeddings=tie_word_embeddings, **kwargs)
+
+
+__all__ = ["LagunaConfig"]

generation_config.json

@@ -0,0 +1,25 @@
+{
+  "bos_token_id": 2,
+  "do_sample": true,
+  "eos_token_id": [
+    2,
+    24
+  ],
+  "max_new_tokens": 32768,
+  "pad_token_id": 9,
+  "temperature": 1.0,
+  "top_p": 1.0,
+  "min_p": 0.0,
+  "top_k": 20,
+  "tool_call_parser": "poolside_v1",
+  "reasoning_parser": "poolside_v1",
+  "default_chat_template_kwargs": {
+    "enable_thinking": true
+  },
+  "speculative_config": {
+    "method": "dflash",
+    "source": "huggingface",
+    "model": "poolside/Laguna-S-2.1-DFlash",
+    "num_speculative_tokens": 15
+  }
+}

model-00001-of-00046.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e8fd968d4c0d1c5ddaccaf93a3bb277d49652f853f628b5c7467c0161632e1dd
+size 5079805456

model-00002-of-00046.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e74b6bb8ba427c87dd8315fadaeb76386405e337f1f0be14fcbd526932e7806f
+size 5118428616

model-00003-of-00046.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:583231c1d7cee5f5c30889d95740ef11f96761fb211c5998c8e8d32dd44ab0e1
+size 5115054984

model-00004-of-00046.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7ce6e48b7aad7ea1d88f4adc2363979268aece2d0cb3451258796f93ba7d3b6c
+size 5115054984

model-00005-of-00046.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:554e0b341efc75449e35373df435b01d1b393897faaaf2b266dfc579c659a738
+size 5115054984

model-00006-of-00046.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:09375e99bde0b699167640f283c6c97630cd04dc8cdf6577099acb6dbbf47132
+size 5115054984

model-00007-of-00046.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:211981662364c537705ab2f001e890a0dad78f317f991cba12328a8430912a20
+size 5115054984

model-00008-of-00046.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:accf003d94c40b9a1b18e1d0af8f09d6283845b0f9ab4d1620237c5acdb4c711
+size 5115054984

model-00009-of-00046.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:beb2b07cd0130203d940bd2d7e111ac3289a2924b987de4a4b66f4a75677c998
+size 5115054984

model-00010-of-00046.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:854c9ecfd640218821382cbe884234aa2ab4a095586b7e79332fe0dd39030451
+size 5115054984

model-00011-of-00046.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7990e4cfcef928737036a79b51dd18b583b97f768191e16a4c9c35ed6a8a90be
+size 5115055728

model-00012-of-00046.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a4499f1834bbb61963e4cab3910d9fa68204e6a79df20e074e33033cc9eccebe
+size 5115055736

model-00013-of-00046.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:21939ed1f057632d2619f8116bef1393f5651f66b42803e11cfb9421272acbe6
+size 5115055744

model-00014-of-00046.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:56a93bce9942193cbc97d1bf0956e6e144d2eda5393f1451a8b29e47c1dd72e3
+size 5115055752

model-00015-of-00046.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8dbe8f40428c9b54dbdbe3bd48bfe24915e7ff6c041dc4f264253cfedf52c131
+size 5115055752

model-00016-of-00046.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c7e321b971707b3202807173286ff9958b1c68d18294a8099c14cd9f19b1e031
+size 5115055752

model-00017-of-00046.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:635d2210c72a5433c54c6648d382907a273113d37565fc8327dfa48467fe4956
+size 5115055752

model-00018-of-00046.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b3cf79b3ec1e02327aa0db2b401695d9f9ad30d554a5a196bc0c664e6efa1814
+size 5115055752

model-00019-of-00046.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:879c8ae5315d6ac85d06af3edd44c17259cc5da41cbb517e209557d8d495a7fe
+size 5115055792

model-00020-of-00046.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b17ade3e3b89f91a837da754c4b5664ca0aaf5ee303853929c2b61e7629e6d7
+size 5115055792

model-00021-of-00046.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3ba6273814850b4f700a1610529d1dd49631077536bf4a647c526cc7757137f0
+size 5115055792

model-00022-of-00046.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:baafcfcaf638eb3281f544f19fe94061947b70b480ccb9fba2ccb1d9c9d375af
+size 5115055792

model-00023-of-00046.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e5887ad7d9449bebd0ca74eb89e66fc365e3de0e2a0957f3129af2a508e425fd
+size 5115055792

model-00024-of-00046.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fd42e9ecc725ca75416220d6928f5d604aa2e930d850e19fc28810076ad46950
+size 5115055792

model-00025-of-00046.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:806fb7ac2acb46e7fa998714346f30ca698203124a104016e96b3cc2afd41914
+size 5115055792

model-00026-of-00046.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:50f5cab6c72b8cefbf73ca198142e91b297063c79999bb9db1d96914004ff2ca
+size 5115055792

model-00027-of-00046.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b136986ef03a9fb0d59b5402431b39a9a6c00e912aa0aa39202313b6f71725b
+size 5115055792

model-00028-of-00046.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b20986e17c2438afae92adacc73d8f0950c3b5fa1df7b841edf6f4599d0aba82
+size 5115055792

model-00029-of-00046.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:06bc11f60a13bd25d5a051b43601174c1eabbfc8a79b7f27ce913f24dc0dd704
+size 5115055744

model-00030-of-00046.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f91f242ba966007796911d413c2a55301c2970e77b6aeb7ae106f969e00143c7
+size 5115055744

model-00031-of-00046.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eadd9d449788573437fd5547ddf1ab47d319bab86d386af09816fdce1a277e56
+size 5115055752

model-00032-of-00046.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3cee82b9e55e10d7dc8e9e6d520307cad812da0578395735d350f2070688b911
+size 5115055752

model-00033-of-00046.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:431608c5d999abb7d221ba7eaea200c15a226f6b70b0745bbf2228bf77244465
+size 5115055752

model-00034-of-00046.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cdccefda3ec897e2120fd6ce61295e76506c3738ff0d411440957e19ec53cdde
+size 5115055752

model-00035-of-00046.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:96b809ce76baa75b689acf44c0b32e55b19668f045a70917cc46218be53c1e12
+size 5115055752

model-00036-of-00046.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2a4768922f56d5882548e93e53afd481b1d4b4090b65f8b7f53ef69acc215b31
+size 5115055784

model-00037-of-00046.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2eb3e2ebb178c804671e03886384a0bf350ac28fa7e5b79c772d2f18818d2d46
+size 5115055792

model-00038-of-00046.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:347c4b5bad0fd5630903ded0cb7b5774563f321e245643dab80e6ca14915a395
+size 5115055792

model-00039-of-00046.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:268a4ce70a6d1dc532100171393f94f0d7c45b9fd0f1fbb538516f20cb1d14ba
+size 5115055792

model-00040-of-00046.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e3ee5dbe50a75da5776ff2e5b94b846e80964a083582696ffbbd3b4912e5050d
+size 5115055792