initial commit

.eval_results/swe-bench_pro.yaml

@@ -0,0 +1,7 @@
+- dataset:
+    id: ScaleAI/SWE-bench_Pro
+    task_id: SWE_Bench_Pro
+  value: 49.2
+  source:
+    url: https://huggingface.co/poolside/Laguna-M.1
+    name: Model Card

.eval_results/swe-bench_verified.yaml

@@ -0,0 +1,7 @@
+- dataset:
+    id: SWE-bench/SWE-bench_Verified
+    task_id: swe_bench_%_resolved
+  value: 74.6
+  source:
+    url: https://huggingface.co/poolside/Laguna-M.1
+    name: Model Card

.eval_results/terminal-bench-2.0.yaml

@@ -0,0 +1,7 @@
+- dataset:
+    id: harborframework/terminal-bench-2.0
+    task_id: terminalbench_2
+  value: 45.8
+  source:
+    url: https://huggingface.co/poolside/Laguna-M.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 +1,202 @@
+
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright 2026 Poolside
+
+   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.

README.md

@@ -0,0 +1,312 @@
+---
+library_name: vllm
+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-m.1
+- vllm
+- bf16
+- moe
+license: apache-2.0
+pipeline_tag: text-generation
+---
+
+<p align="center">
+  <img alt="poolside-banner" src="https://poolside.ai/assets/laguna/laguna-m1-banner.svg" width="800px">
+</p>
+
+<p align="center">
+  <a href="https://platform.poolside.ai"><strong>Get an API key</strong></a> ·
+  <a href="https://poolside.ai/blog/laguna-a-deeper-dive"><strong>Release blog post</strong></a> ·
+  <a href="https://poolside.ai/assets/laguna/laguna-m1-xs2-technical-report.pdf"><strong>Technical report</strong></a>
+</p>
+
+<br>
+
+# Laguna M.1
+
+Laguna M.1 is a 225B total parameter Mixture-of-Experts model with 23B activated parameters per token designed for agentic coding and long-horizon work. This release has upstream support in vLLM and is part of stable releases from version v0.21.0.
+
+> [!NOTE]
+> For more details on how we trained this model, including our Model Factory approach, post-training recipe, async off-policy agent RL, and evaluations, check out our [release blog post](https://poolside.ai/blog/laguna-a-deeper-dive) and [technical report](https://poolside.ai/assets/laguna/laguna-m1-xs2-technical-report.pdf).
+
+## Highlights
+
+* **Large sparse MoE for agentic coding**: Laguna M.1 is a 70-layer MoE transformer with 225B total parameters and 23B activated parameters per token
+* **High-capacity expert routing**: After 3 dense SwiGLU layers, Laguna M.1 uses 67 sparse MoE layers with 256 experts, top-k=16 routing and auxiliary-loss-free load balancing
+* **Global attention architecture**: Laguna M.1 uses global attention across all layers with 64 Q-heads, 8 KV-heads and softplus attention output gating
+* **Native reasoning support**: Interleaved thinking between tool calls with support for enabling and disabling thinking per-request
+* **Strong agentic benchmark performance**: Laguna M.1 is competitive with state-of-the-art open-weight and frontier models on SWE-bench Verified, SWE-bench Multilingual, SWE-Bench Pro and Terminal-Bench 2.0
+* **Apache 2.0 license**: Use and modify freely for commercial and non-commercial purposes
+
+---
+
+## Model overview
+
+- Training: pre-training, post-training and reinforcement learning stages
+- Number of parameters: 225B total with 23B activated per token
+- Optimizer: Muon
+- Layers: 70 layers with global attention
+- Experts: 256 experts with 1 shared expert; top-k=16 routing
+- Dense layers: first 3 layers are dense SwiGLU; remaining 67 layers are sparse MoE
+- Attention: 64 Q-heads, 8 KV-heads, head dimension 128, with softplus attention output gating
+- Positional encoding: RoPE with YaRN
+- Modality: text-to-text
+- Context window: 262,144 tokens
+- Reasoning support: interleaved thinking with preserved thinking
+
+## Benchmark results
+
+<p align="center">
+  <img alt="benchmarks" src="https://poolside.ai/assets/laguna/laguna-m1-chart.svg" width="800px">
+</p>
+
+| Model                     | Parameters           | SWE-bench Verified | SWE-bench Multilingual | SWE-bench Pro (Public Dataset) | Terminal-Bench 2.0 |
+|---------------------------|----------------------|--------------------|------------------------|--------------------------------|--------------------|
+| **Laguna M.1**            | 225B-A23B            | 74.6%              | 63.1%                  | 49.2%                          | 45.8%              |
+| Devstral 2                | 123B dense           | 72.2%              | 61.3%                  | -                              | 32.6%              |
+| GLM-4.7                   | 355B-A32B            | 73.8%              | 66.7%                  | -                              | 41.0%              |
+| DeepSeek-V4 Flash         | 284B-A13B            | 79.0%              | 73.3%                  | 52.6%                          | 56.9%              |
+| Qwen3.5-397B-A17B         | 397B-A17B            | 76.2%              | 69.3%                  | 50.9%                          | 52.5%              |
+| Claude Sonnet 4.6         | -                    | 79.6%              | -                      | -                              | 59.1%              |
+
+*We used the highest publicly-referenced scores for all comparison models across each benchmark. In almost all cases these were official scores published in release blog posts or equivalent, with Claude Sonnet 4.6 shown as a frontier proprietary reference of comparable model size. “-” indicates a score not reported by the model provider.*
+
+> [!NOTE]
+> All benchmarking for Laguna M.1 was completed using our [pool agent harness](https://github.com/poolsideai/pool), with a maximum of 500 steps and sandboxed execution. The same sampling parameters were used for all Laguna M.1 benchmarking: temperature=1.0 and top_k=20, with thinking mode enabled and a context length of 256K tokens. All tasks were run in their own sandbox using 8 GB RAM/2 CPUs, with the exception of Terminal-Bench 2.0, which used 48 GB RAM/32 CPUs.
+>
+> Some base task images and verifiers were patched to fix infrastructure reliability issues inherent in task setup, such as rate limits on third-party dependencies in external registries used by the verifier. All four agentic benchmarks were run with patched images. We also ran a reward-hack judge post-hoc on Laguna M.1 evaluation runs and did not find significant reward hacking after joint judge review and manual review.
+>
+> - SWE-bench Verified: mean pass@1 averaged over 4 runs
+> - SWE-bench Multilingual: mean pass@1 averaged over 4 runs
+> - SWE-Bench Pro: mean pass@1 averaged over 4 runs
+> - Terminal-Bench 2.0: mean pass@1 averaged over 4 runs; 48 GB RAM/32 CPUs
+
+## Usage
+
+Laguna M.1 has upstream support in vLLM and Transformers, and TRT-LLM thanks to the support of the team at NVIDIA.
+
+### pool
+
+**pool** is a lightweight terminal-based coding agent and a dual [Agent Client Protocol](https://agentclientprotocol.com/get-started) client-server.
+
+Download and install for macOS and Linux:
+
+```shell
+curl -fsSL https://downloads.poolside.ai/pool/install.sh | bash
+```
+
+Launch and *Log in with Poolside* to get a free API key.
+
+```shell
+pool
+```
+
+Use in any [ACP client](https://agentclientprotocol.com/get-started/clients). Configure Zed and JetBrains automatically:
+
+```shell
+pool acp setup --editor zed|jetbrains
+```
+
+#### Feedback and issues
+
+Submit feedback with `/feedback` and read the [full documentation on GitHub](https://github.com/poolsideai/pool).
+
+### Deployment
+
+#### vLLM
+
+Serve Laguna M.1 locally with vLLM and query it from any OpenAI-compatible client (see [Controlling reasoning](#controlling-reasoning) for tool calls, streaming, and reasoning extraction):
+
+> [!NOTE]
+> Laguna support landed in vLLM via [vllm-project/vllm#41129](https://github.com/vllm-project/vllm/pull/41129) (shared with [Laguna XS.2](https://huggingface.co/poolside/Laguna-XS.2)) and is available in vLLM 0.21.0 and later.
+
+```shell
+pip install 'vllm>=0.21.0'
+
+vllm serve \
+    --model poolside/Laguna-M.1 \
+    --tool-call-parser poolside_v1 \
+    --reasoning-parser poolside_v1 \
+    --enable-auto-tool-choice \
+    --served-model-name laguna \
+    --default-chat-template-kwargs '{"enable_thinking": true}'
+```
+
+See the [vLLM recipes page](https://recipes.vllm.ai/poolside/Laguna-XS.2) for our Laguna XS.2 model with which the implementation is shared for additional deployment guidance. FP8 and NVFP4 quantized checkpoints are available at [Laguna-M.1-FP8](https://huggingface.co/poolside/Laguna-M.1-FP8) and [Laguna-M.1-NVFP4](https://huggingface.co/poolside/Laguna-M.1-NVFP4); quantization is detected automatically from `quantization_config`, so the same command works with the model ID substituted.
+
+#### Transformers
+
+Laguna is supported in Transformers `v5.7.0` and later ([huggingface/transformers#45673](https://github.com/huggingface/transformers/pull/45673)).
+
+> [!NOTE]
+> Laguna M.1 is a 225B-parameter model; loading the BF16 checkpoint in Transformers requires substantial multi-GPU memory (`device_map="auto"` shards across available devices). For single-node serving, vLLM is recommended.
+
+```python
+import torch
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+model_id = "poolside/Laguna-M.1"
+
+tokenizer = AutoTokenizer.from_pretrained(model_id)
+model = AutoModelForCausalLM.from_pretrained(
+    model_id,
+    dtype=torch.bfloat16,
+    device_map="auto",
+)
+
+messages = [
+    {"role": "user", "content": "Write a Python retry wrapper with exponential backoff."},
+]
+
+# Reasoning is on by default; pass enable_thinking=False to skip the <think> block.
+inputs = tokenizer.apply_chat_template(
+    messages,
+    add_generation_prompt=True,
+    return_tensors="pt",
+    enable_thinking=True,
+).to(model.device)
+
+outputs = model.generate(inputs, max_new_tokens=1024, do_sample=True, temperature=1.0, top_k=20)
+print(tokenizer.decode(outputs[0][inputs.shape[-1]:], skip_special_tokens=True))
+```
+
+#### TRT-LLM
+
+Laguna is supported in TensorRT-LLM thanks to the team at NVIDIA — model support landed in [NVIDIA/TensorRT-LLM#13559](https://github.com/NVIDIA/TensorRT-LLM/pull/13559), with partial-RoPE fusion added in [#15110](https://github.com/NVIDIA/TensorRT-LLM/pull/15110). Build TensorRT-LLM from a `main` that includes these PRs (or a release once they ship).
+
+```python
+from tensorrt_llm import LLM, SamplingParams
+
+llm = LLM(model="poolside/Laguna-M.1", trust_remote_code=True)
+sampling = SamplingParams(max_tokens=1024, temperature=1.0, top_k=20)
+out = llm.generate(["Write a Python retry wrapper with exponential backoff."], sampling)
+print(out[0].outputs[0].text)
+```
+
+> [!NOTE]
+> If your TensorRT-LLM build pins `transformers < 4.58`, `configuration_laguna.py` needs a small compat shim; use the `laguna_minimal_overlay.sh` helper from the support PR and load TRT-LLM against the overlay directory.
+
+Quantization is detected automatically from `quantization_config`, so the same recipe works for the [FP8](https://huggingface.co/poolside/Laguna-M.1-FP8) and [NVFP4](https://huggingface.co/poolside/Laguna-M.1-NVFP4) variants with no extra flags.
+
+## Controlling reasoning
+
+Laguna M.1 has native reasoning support and is designed to work best with *preserved thinking*, where `reasoning` content from prior assistant messages is preserved in the message history. This model will generally reason before calling tools and between tool calls.
+
+```python
+import json
+from openai import OpenAI
+
+client = OpenAI(
+  base_url="https://inference.poolside.ai/v1",
+  api_key="...",
+)
+
+model = "poolside/laguna-m.1"
+
+tools = [{"type": "function", "function": {
+  "name": "shell",
+  "description": "Execute a bash command and return the output.",
+  "parameters": {"type": "object", "properties": {"cmd": {"type": "string"}}, "required": ["cmd"]},
+}}]
+
+messages = [
+  {"role": "system", "content": "You are a coding agent with access to a shell tool."},
+  {"role": "user", "content": "Run uname -a"},
+]
+
+# Thinking is enabled by default when the server sets --default-chat-template-kwargs {"enable_thinking": True}
+# When using the Poolside API (https://inference.poolside.ai/v1), this flag is set by default
+response = client.chat.completions.create(
+  model=model,
+  messages=messages,
+  tools=tools,
+  stream=True,
+)
+
+reasoning, content, tool_calls = "", "", []
+for chunk in response:
+  delta = chunk.choices[0].delta
+  if hasattr(delta, "reasoning_content") and delta.reasoning_content:
+    reasoning += delta.reasoning_content
+  if hasattr(delta, "content") and delta.content:
+    content += delta.content
+  if hasattr(delta, "tool_calls") and delta.tool_calls:
+    for tc in delta.tool_calls:
+      if tc.index >= len(tool_calls):
+        tool_calls.append({"id": tc.id, "function": {"name": "", "arguments": ""}})
+      if tc.function.name:
+        tool_calls[tc.index]["function"]["name"] = tc.function.name
+      if tc.function.arguments:
+        tool_calls[tc.index]["function"]["arguments"] += tc.function.arguments
+
+print(f"Reasoning: {reasoning}\nContent: {content}\nTool calls: {tool_calls}\n")
+
+# Return reasoning in the next request for best performance
+messages.append({
+  "role": "assistant",
+  "content": content,
+  "reasoning_content": reasoning,
+  "tool_calls": [{"id": tc["id"], "type": "function", "function": tc["function"]} for tc in tool_calls]
+})
+
+messages.append({
+  "role": "tool",
+  "tool_call_id": tool_calls[0]["id"],
+  "content": json.dumps({"stdout": "Darwin arm64", "exit_code": "0"})
+})
+
+response = client.chat.completions.create(
+  model=model,
+  messages=messages,
+  tools=tools,
+  stream=True,
+)
+
+reasoning, content = "", ""
+for chunk in response:
+  delta = chunk.choices[0].delta
+  if hasattr(delta, "reasoning_content") and delta.reasoning_content:
+    reasoning += delta.reasoning_content
+  if hasattr(delta, "content") and delta.content:
+    content += delta.content
+
+print(f"Reasoning: {reasoning}\nContent: {content}")
+```
+
+### Disabling reasoning
+
+You can disable thinking by setting `enable_thinking` to `False` in a request or by not providing `--default-chat-template-kwargs {"enable_thinking": True}` or equivalent when starting the server.
+
+```python
+from openai import OpenAI
+client = OpenAI()
+
+completion = client.chat.completions.create(
+  model="poolside/laguna-m.1",
+  messages=[
+    {"role": "user", "content": "Write a retry wrapper with exponential backoff."}
+  ],
+  extra_body={
+    "chat_template_kwargs": { "enable_thinking": False },
+  },
+  stream=True
+)
+
+for chunk in completion:
+    print(chunk.choices[0].delta)
+```
+
+For agentic coding use cases, we recommend enabling thinking and preserving reasoning in message history as outlined in the [Controlling reasoning](#controlling-reasoning) section.
+
+## License
+
+This model is licensed under the [Apache 2.0 License](https://huggingface.co/poolside/Laguna-M.1/blob/main/LICENSE.md).
+
+## Intended and Responsible Use 
+
+Laguna M.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 M.1 is subject to the [Apache 2.0 License](https://huggingface.co/poolside/Laguna-M.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 M.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 +1,132 @@
+{#- Copied from laguna_glm_thinking_v4/chat_template.jinja -#}
+{#- Removes prefix that references <think> token, and replaces message.reasoning_content reference with message.reasoning -#}
+{{- "〈|EOS|〉" -}}
+{%- set enable_thinking = enable_thinking | default(false) -%}
+{%- set render_assistant_messages_raw = render_assistant_messages_raw | default(false) -%}
+{%- set add_generation_prompt = add_generation_prompt | default(false) -%}
+
+{#- ───── header (system message) ───── -#}
+{%- set system_message = "" -%}
+{%- if messages and messages[0].role == "system" -%}
+  {%- set system_message = messages[0].content -%}
+{%- endif -%}
+
+{%- if (system_message and system_message.strip()) or tools -%}
+  {{- "<system>\n" -}}
+
+  {%- if system_message and system_message.strip() -%}
+    {{- "\n" -}}
+    {{- system_message.rstrip() -}}
+  {%- endif -%}
+
+  {%- if tools -%}
+    {{- "\n\n### Tools\n\n" -}}
+    {%- set ns = namespace(tool_string="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 -%}
+      {%- set ns.tool_string = ns.tool_string ~ (tool | tojson) ~ "\n" -%}
+    {%- endfor -%}
+    {%- if enable_thinking -%}
+      {%- set tool_string = ns.tool_string + "</available_tools>\n\n" ~
+            "Wrap your thinking in '<think>', '</think>' tags, followed by a function call. For each function call, return an unescaped XML-like object with function name and arguments within '<tool_call>' and '</tool_call>' tags, like here:\n" ~
+            "<think> your thoughts here </think>\n" ~
+            "<tool_call>function-name\n<arg_key>argument-key</arg_key>\n<arg_value>value-of-argument-key</arg_value>\n" ~
+            "</tool_call>" -%}
+    {%- else -%}
+      {%- set tool_string = ns.tool_string + "</available_tools>\n\n" ~
+            "For each function call, return an unescaped XML-like object " ~
+            "with function name and arguments within '<tool_call>' and '</tool_call>' tags, like here:\n" ~
+            "<tool_call>function-name\n<arg_key>argument-key</arg_key>\n<arg_value>value-of-argument-key</arg_value>\n" ~
+            "</tool_call>" -%}
+    {%- endif -%}
+    {{- tool_string -}}
+  {%- endif -%}
+
+  {{- "\n</system>\n" -}}
+{%- endif -%}
+
+{#- ───── main loop ───── -#}
+{%- for message in messages -%}
+  {%- set content = message.content if message.content is string else "" -%}
+  {%- if message.role == "user" -%}
+    {{- "<user>\n" + content + "\n</user>\n" -}}
+  {%- elif message.role == "assistant" -%}
+    {%- generation -%}
+      {{- "<assistant>\n" -}}
+      {%- if render_assistant_messages_raw -%}
+        {#- Raw mode: prepend the generation prompt token, then dump content verbatim. -#}
+        {#- The generation prompt is <think> when enable_thinking, </think> otherwise. -#}
+        {#- Only prepend if content doesn't already start with it. -#}
+        {%- if enable_thinking -%}
+          {%- if not content.startswith('<think>') -%}
+            {{- '<think>' -}}
+          {%- endif -%}
+        {%- else -%}
+          {%- if not content.startswith('</think>') -%}
+            {{- '</think>' -}}
+          {%- endif -%}
+        {%- endif -%}
+        {{- content -}}
+        {#- Append closing tag if content doesn't already end with it. -#}
+        {%- if not content.endswith('</assistant>\n') and not content.endswith('</assistant>') -%}
+          {{- '\n</assistant>' -}}
+        {%- endif -%}
+        {{- "\n" -}}
+      {%- else -%}
+        {#- Extract reasoning content from message.reasoning (vLLM field name) or message.reasoning_content, or from <think> tags -#}
+        {%- 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 %}
+        {#- Always strip <think> tags from content if present to avoid duplication -#}
+        {%- if '</think>' in content %}
+          {%- if not reasoning_content %}
+            {%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
+          {%- endif %}
+          {%- set content = content.split('</think>')[-1].lstrip('\n') %}
+        {%- endif %}
+        {#- Display reasoning content for all messages -#}
+        {%- if reasoning_content -%}
+          {{- '<think>\n' + reasoning_content.strip() + '\n</think>\n' -}}
+        {%- else -%}
+          {{- '</think>\n' -}}
+        {%- endif -%}
+        {#- Display main content -#}
+        {%- if content.strip() -%}
+          {{- content.strip() ~ "\n" -}}
+        {%- 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>\n" -}}
+              {{- "<arg_value>"}}{{ v | tojson(ensure_ascii=False) if v is not string else v }}{{ "</arg_value>\n" -}}
+            {%- endfor -%}
+            {{- "</tool_call>\n" -}}
+          {%- endfor -%}
+        {%- endif -%}
+        {{- "</assistant>\n" -}}
+      {%- endif -%}
+    {%- endgeneration -%}
+  {%- elif message.role == "tool" -%}
+    {{- "<tool_response>\n" + content + "\n</tool_response>\n" -}}
+  {%- elif message.role == "system" and loop.index0 != 0 -%}
+    {#- Render additional system messages (skip the first one which is handled separately in the header) -#}
+    {{- "<system>\n" + content + "\n</system>\n" -}}
+  {%- endif -%}
+{%- endfor -%}
+{#- ───── generation prompt ───── -#}
+{%- if add_generation_prompt -%}
+  {{- "<assistant>\n" -}}
+  {#- ───── Include reasoning mode directive ───── -#}
+  {%- if not enable_thinking %}
+    {{- '</think>' -}}
+  {%- else %}
+    {{- '<think>' -}}
+  {%- endif %}
+{%- endif -%}

config.json

@@ -0,0 +1,202 @@
+{
+  "architectures": [
+    "LagunaForCausalLM"
+  ],
+  "auto_map": {
+    "AutoConfig": "configuration_laguna.LagunaConfig",
+    "AutoModelForCausalLM": "modeling_laguna.LagunaForCausalLM"
+  },
+  "model_type": "laguna",
+  "vocab_size": 100352,
+  "hidden_size": 4096,
+  "intermediate_size": 16384,
+  "num_hidden_layers": 70,
+  "num_attention_heads": 64,
+  "num_key_value_heads": 8,
+  "head_dim": 128,
+  "max_position_embeddings": 131072,
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "rms_norm_eps": 1e-06,
+  "num_experts": 256,
+  "num_experts_per_tok": 16,
+  "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,
+    1,
+    2
+  ],
+  "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-element",
+  "sliding_window": 0,
+  "rope_parameters": {
+    "full_attention": {
+      "rope_theta": 500000.0,
+      "rope_type": "yarn",
+      "factor": 32.0,
+      "original_max_position_embeddings": 4096,
+      "beta_slow": 1.0,
+      "beta_fast": 64.0,
+      "attention_factor": 1.0,
+      "partial_rotary_factor": 1.0
+    }
+  },
+  "moe_apply_router_weight_on_input": false,
+  "mlp_layer_types": [
+    "dense",
+    "dense",
+    "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",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse",
+    "sparse"
+  ],
+  "gating_types": [
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element",
+    "per_element"
+  ],
+  "moe_routed_scaling_factor": 1.0
+}

configuration_laguna.py

@@ -0,0 +1,245 @@
+# 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}
+
+        # 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 +1,13 @@
+{
+  "bos_token_id": 2,
+  "do_sample": true,
+  "eos_token_id": [
+    2,
+    24
+  ],
+  "max_new_tokens": 4096,
+  "pad_token_id": 9,
+  "temperature": 1.0,
+  "top_p": 1.0,
+  "min_p": 0.0
+}

model-00001-of-00089.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:230bfc34d0a67f0b92d8beb3dcc94f628ebe82d82c1327f203dab9e0e99b121c
+size 5119451664

model-00002-of-00089.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4bbeb3ecbfa97602d8bd9fab4f078ae33df3867d94df529cf12ab875058c5e55
+size 5085993864

model-00003-of-00089.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9b34bd3857c3b6872790482145fe3b2bb81476e81a05cf5b4ebdc6b2d081950b
+size 5117432672

model-00004-of-00089.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:230f8d942b97162d0633373db2e5d33b73172fa1dda3ce65ec477ba95aa5fd2a
+size 5119410552

model-00005-of-00089.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4a052df48579439f9c2fdd0101d62d6fc6514ac21f7e70d6dfd86b4e06f4c33d
+size 5117126864

model-00006-of-00089.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:763bb266a50c6303ded30cd823b3adf43f1bcd5ee1909add6df9a0bbe5f75c2a
+size 5117126776

model-00007-of-00089.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7104c3c1f8046c56d19737b709479a690f22f62d5edc64b90dc13a769c8b5de1
+size 5117126776

model-00008-of-00089.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a800e72938eb992cd81086b76dcf112bea1243cb4095e4bf8785242d1ace1a81
+size 5117126880

model-00009-of-00089.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7c69db1555b4499353d52b45d324ad419a5fd400fe2c81bb5398286f26f4bdb0
+size 5117126936

model-00010-of-00089.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:848011466cc779a33ea374925d7d1866842eaa2743cc3e3eaa34795b63c7a3e7
+size 5117126840

model-00011-of-00089.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:60252cde42da7708216f8f8b9e85031c6d5a2e54f3f76baf06833a5a62346b9c
+size 5117126776

model-00012-of-00089.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4bd9e973e6825554a5364330b62f0cc99aecbeab4da8cbaba33174eb09d258a3
+size 5117126776

model-00013-of-00089.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ef3b1a12eb31a3ad70b3bd369bce9e27a483c3e7796d7a3379cf6435f4be5ab6
+size 5117127072

model-00014-of-00089.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:94900c9bacf4bf50cd598d1e3145fd28d0ed610cf7e986895bd786c68a9debec
+size 5117127560

model-00015-of-00089.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:52f0f996d77d7e5fe1a84da4c21df6cb682d7a77f36d22be08dbb0eb8b412fb3
+size 5117127416

model-00016-of-00089.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:399fd29b77e8b22c2c9a74ab3b6f2475a1535a38af39fb16cc83bde2610a35c1
+size 5117127384

model-00017-of-00089.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dbf889d680b1598107470ba966be9e47593b5cf97c93b7eafe1848bae18f4f81
+size 5117127384

model-00018-of-00089.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2aab9519c42b1fc59bc072376d26c5a6f1df59ed957c0b9729fdf7b9a56caf20
+size 5117127536

model-00019-of-00089.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6ae508d7070a9506a6765361b69b51198a1a73194dd23de2558ba8167a3aaa52
+size 5117127568

model-00020-of-00089.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:42a9eec6339221f19de1e17287edc9965d9a94b73ccd0ba380743871cc570ee7
+size 5117127384

model-00021-of-00089.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e8bdf3c1be54a60fefb13f9b0d1dec708ca71a13633a8c55bce21886b52a0319
+size 5117127384

model-00022-of-00089.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2b4b6db66886d150c907216b2fdf152a4bd220f24b485e4e0e1599281dde24cb
+size 5117127400

model-00023-of-00089.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fd05b9a592c81a32761a93a075d69c3a49aff1a538c27325044a634d89da906a
+size 5117127544

model-00024-of-00089.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b7df12b9b155563d10e32ab9989171380a16e4859be334c55da4a705053c80a3
+size 5117127544

model-00025-of-00089.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c0134632cc7f7db2e77f6ec7ae4b953c2ec1af5de4b1040ff2eb31bcb0ab8dd
+size 5117127384

model-00026-of-00089.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ccaef3e08e56d331aa627231e2e0e7bc90b861d5388a6529b6a4604e7afe58a7
+size 5117127384

model-00027-of-00089.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:90d16c2444c867f041049be6e5946d5e786588bcba1bf50394c3d7fa4461163e
+size 5117127424

model-00028-of-00089.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:72083a61686e7f2d52d3579bd60a43788fdfbddf1e6d41e403fee850200d19cb
+size 5117127544

model-00029-of-00089.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bfe01af6c42e2366d0eb8629c40ec2674df839a16f5b81be37ec2cbfc2906a12
+size 5117127520

model-00030-of-00089.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3aea6291d6c5ccfd143d38f4b8dc2b3f1f3cc76e735ed8a352e1d32f956ddcc2
+size 5117127384

model-00031-of-00089.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:98243efb22ac856c50215c4cea159da7210d24f54be0a31d182f370e542dc15a
+size 5117127384

model-00032-of-00089.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d0e861f320f8aef0c692c28c602fe92e0fd9d57b965b9623cacf3e06cc651f54
+size 5117127440

model-00033-of-00089.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c4e612640547e1691451884b6bbbe64a30c0835d7c9846f14b4042b7280c32ae
+size 5117127544

model-00034-of-00089.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ae308c049b026e630cff7ce1fedd73e522c79c234814b0d4ef0b3cf7d00bbdc4
+size 5117127496

model-00035-of-00089.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ba5bff98b7ea4fe119d8ebb252a18459098bd13853febca12d6efbccbd97213e
+size 5117127384

model-00036-of-00089.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b5142bc53a51a456bf12e38cc4e994b50a03ff28335068f5adfb9430d24e500e
+size 5117127384

model-00037-of-00089.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9148cd551bb4611c168ed1cb46877e48c8c7f4157f71889e2872b29c1283d5b5
+size 5117127464

model-00038-of-00089.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ecb9aa08e087fef94805bcc2f91f8277fe5b19f3a7b2b01a680d90c7dd0b691e
+size 5117127544

model-00039-of-00089.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:efe972d5de11a0b0f218a52400fc8a325e835b6a4746c7c6a91d4ae4887c422e
+size 5117127480

model-00040-of-00089.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61b6b73c68a6d678cbf424e8a25d50d1f118fa1ead62351e48de16978e657c47
+size 5117127384