Duplicate from CohereLabs/North-Mini-Code-1.0

Co-authored-by: Alejandro Rodriguez <alexrs@users.noreply.huggingface.co>

.gitattributes

@@ -0,0 +1,36 @@
+*.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
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,277 @@
+---
+inference: false
+library_name: transformers
+license: apache-2.0
+tags:
+- conversational
+- chat
+- code
+- agent
+---
+
+# **Model Card for North Mini Code**
+
+## **Model Summary**
+
+North Mini Code is an open weights research release of a 30B-A3B parameter model optimized for code generation, agentic software engineering, and terminal tasks.
+
+Developed by: [Cohere](https://cohere.com/) and [Cohere Labs](https://cohere.com/research)
+
+* Point of Contact: [**Cohere Labs**](https://cohere.com/research)  
+* License: Apache 2.0
+* Model: North Mini Code  
+* Model Size: 30B total; 3B active  
+* Context length: 256K & 64K max output
+
+For more details about this model, please check out our [blog post](https://huggingface.co/blog/CohereLabs/introducing-north-mini-code). 
+
+**Try North Mini Code**
+
+You can try out North Mini Code before downloading the weights in OpenCode and our hosted [Hugging Face Space](https://huggingface.co/spaces/CohereLabs/North-Mini-Code-1.0).
+
+**Evaluation**
+
+![image1](https://cdn-uploads.huggingface.co/production/uploads/62668f725fb8d521d94d8451/xR7kZ3X9RKEZrbgD6hpG1.png)
+
+<details>
+<summary><span style="font-size: 80%;"><b>Benchmarking Methodology [CLICK TO EXPAND]</span></b></summary>
+
+- <span style="font-size: 80%;">We used SWE-Bench Verified, SWE-Bench Pro, Terminal-Bench v2, and Terminal-Bench Hard to benchmark North Mini Code's agentic coding capabilities. For evaluation harnesses, we used the Swe-Agent harness v1.1.0 for SWE-Bench, and a simple ReAct harness employing a single terminal-use tool based on Harbor's Tmux session implementation for Terminal-Bench v2. For Terminal Bench Hard, we directly used Terminus-2, following the same methodology as the Artificial Analysis Intelligence Index to compare North-Mini-Code-1.0 with the other models. Additionally, we used SciCode and LiveCodeBench v6 as complex code-generation benchmarks outside of tool use.</span>
+- <span style="font-size: 80%;">We run each benchmark with 3 different seeds and report the average benchmark performance, using temperature=1.0 and top\_p=0.95. We used publicly reported scores for competitor models, either from original reports or the Artificial Analysis Intelligence Index, where available. Additionally, Gemma4’s scores for agentic coding tasks were reported by [Qwen team](https://qwen.ai/blog?id=qwen3.6-35b-a3b). For benchmark results that any public report is missing, denoted by (\*) in the figure, we run them internally using the recommended model configuration.</span>
+</details>
+
+**Usage**
+
+Please install transformers from the source repository that includes the necessary changes for this model. We recommend using the following set of sampling parameters for generation: \`temperature=1.0\`, \`top\_p=0.95\`.
+
+```py
+# pip install transformers
+from transformers import AutoTokenizer, AutoModelForCausalLM
+
+model_id = "CohereLabs/North-Mini-Code-1.0"
+tokenizer = AutoTokenizer.from_pretrained(model_id)
+model = AutoModelForCausalLM.from_pretrained(model_id)
+
+prompt = "Write a python program to check if a string is a palindrome or not."
+
+# Format message with the North-Mini-Code-1.0 chat template
+messages = [{"role": "user", "content": prompt}]
+input_ids = tokenizer.apply_chat_template(
+    messages,
+    tokenize=True,
+    add_generation_prompt=True,
+    return_tensors="pt",
+)
+
+gen_tokens = model.generate(
+    **input_ids, 
+    max_new_tokens=1024, 
+    do_sample=True, 
+    temperature=1.0,
+    top_p=0.95
+)
+
+gen_text = tokenizer.decode(gen_tokens[0])
+print(gen_text)
+```
+
+You can also use the model directly using transformers `pipeline` abstraction:
+
+```py
+from transformers import pipeline
+import torch
+
+model_id = "CohereLabs/North-Mini-Code-1.0"
+
+prompt = """Given a list of unique words each of size k and an n sized word, w, where n is a multiple of k,
+Write a program in python to determine the number of unique combinations of words in the list that can be concatenated to form an anagram of the word w.
+"""
+
+pipe = pipeline(
+    "text-generation",
+    model=model_id,
+    torch_dtype="auto",
+    device_map="auto",
+)
+
+messages = [
+    {"role": "user", "content": f"{prompt}"},
+]
+
+text = tokenizer.apply_chat_template(
+    messages,
+    tokenize=False,
+    add_generation_prompt=True,
+)
+
+
+outputs = pipe(
+    messages,
+    max_new_tokens=1024,
+    do_sample=True, 
+    temperature=1.0,
+    top_p=0.95
+
+)
+
+print(outputs[0]["generated_text"][-1])
+
+```
+
+## **Model Details**
+
+**Input**: Text only.
+
+**Output**: Model generates text. 
+
+**Model Architecture**: North-Mini-Code-1.0 is a decoder-only Transformer-based sparse Mixture-of-Experts model. It uses an efficient attention implementation, interleaved between sliding-window attention with RoPE and global attention with no positional embeddings, in a 3:1 ratio. The feed-forward block is an MoE block with 128 experts, of which 8 are activated per token. Each expert block is an FFN block with SwiGLU activation. The router applies a sigmoid activation function to the logits before the top-k selection. We also use a single dense layer before the sparse layers. North-Mini-Code-1.0 was post-trained using a two-stage cascaded supervised fine-tuning (SFT) followed by reinforcement learning with verifiable rewards (RLVR), focusing on agentic coding.  For more technical details, please check out our [blog post](https://huggingface.co/blog/CohereLabs/introducing-north-mini-code). 
+
+**Context Length:** North-Mini-Code-1.0 supports a context length of 256K & 64K output length.
+
+### **Tool Use Capabilities:**
+
+North-Mini-Code-1.0 has been specifically trained with tool-use capabilities for agentic coding.
+
+Tool use with North-Mini-Code-1.0 is supported through [chat templates](https://huggingface.co/docs/transformers/main/en/chat_templating#advanced-tool-use--function-calling) in Transformers. We recommend providing tool descriptions using JSON schema.
+
+**Tool Use Example \[CLICK TO EXPAND\]** 
+
+```py
+# Define tools
+tools = [{
+  "type": "function",
+  "function": {
+    "name": "bash",
+    "description": "Execute a bash command in the terminal.",
+    "parameters": {
+      "type": "object",
+      "properties": {
+        "command": {
+          "description": "The bash command to execute.",
+          "type": "string"
+        }
+      },
+      "required": ["command"]
+    },
+  }
+}]
+
+# Define conversation input
+conversation = [{"role": "user", "content": "Find out if there is any json file in this folder"}]
+
+
+# Get the Tool Use prompt
+input_prompt = tokenizer.apply_chat_template(conversation=conversation, tools=tools, tokenize=False, add_generation_prompt=True, return_tensors="pt")
+
+# Tokenize the prompt
+input_ids = tokenizer(input_prompt, return_tensors="pt")
+```
+
+You can then generate from this input as normal.
+
+North Mini Code, similarly as all the other Cohere agent models released to date, supports [interleaved thinking](https://docs.vllm.ai/en/latest/features/interleaved_thinking/) and works best when turned on. You’re strongly encouraged to pass on all the model-generated thinking contents to future agentic steps, and chat turns for the best model performance. Please refer to the linked vllm doc and see how it’s done.
+
+If the model generates thinking content and tool calls, you should add both of them to the chat history like so:
+
+```py
+# Pass on the tool_call and thinking
+tool_call = {"name": "bash", "arguments": {"command": "ls -al"}}
+reasoning = "The user wants to find if there are any JSON files in the current folder. I should use the `ls` command to list files and then check if there are any JSON files (files ending with .json). Let me first list the files in the current directory."
+
+conversation.append({"role": "assistant", "tool_calls": [{"id": "0", "type": "function", "function": tool_call}], "reasoning": reasoning})
+```
+
+and then call the tool and append the result, as a dictionary, with the tool role, like so:
+
+```py
+# This needs to be a dictionary
+tool_result = {"stdout": "test.json\ntest.py", "return_code": "0"} 
+
+# Append tool results
+conversation.append({"role": "tool", "tool_call_id": "0", "content": tool_result})
+```
+
+After that, you can `generate()` again to let the model use the tool result in the chat.
+
+Note that this was a very brief introduction to tool calling \- for more information the Transformers [tool use documentation](https://huggingface.co/docs/transformers/main/chat_templating#advanced-tool-use--function-calling).
+
+### **vLLM**
+
+You can also run the model in vLLM. Please use vLLM main for North Mini Code until a new release is available, and accurate response parsing also requires installing Cohere’s melody library.
+
+```shell
+uv pip install "git+https://github.com/vllm-project/vllm.git"
+uv pip install cohere_melody>=0.9.0
+```
+
+Then the vllm server can be started with the following command:
+
+```shell
+vllm serve CohereLabs/North-Mini-Code-1.0 \
+  -tp 2 \
+  --max-model-len 320000 \
+  --tool-call-parser cohere_command4 \
+  --reasoning-parser cohere_command4 \
+  --enable-auto-tool-choice
+```
+
+**Use locally deployed North Mini Code in OpenCode:**
+
+Please use OpenCode main branch until a new release is available.
+
+```shell
+# Example commands to install on linux
+git clone https://github.com/anomalyco/opencode.gitcd opencode
+
+# Install Bun
+curl -fsSL https://bun.sh/install | bash
+export BUN_INSTALL="$HOME/.bun"
+export PATH="$BUN_INSTALL/bin:$PATH"
+
+# node-gyp was needed by a dependency
+bun add -g node-gyp
+
+# Install dependencies
+bun install
+
+# Build CLI
+bun run --cwd packages/opencode build/usr/bin/install -m 755 \
+  ./opencode/packages/opencode/dist/opencode-linux-x64/bin/opencode \
+  /root/.local/bin/opencode
+```
+
+To use locally deployed North Mini Code in Opencode, please use this config which enables interleaved reasoning:
+
+```json
+{
+  "$schema": "https://opencode.ai/config.json",
+  "model": "vllm/CohereLabs/North-Mini-Code-1.0",
+  "provider": {
+    "vllm": {
+      "npm": "@ai-sdk/openai-compatible",
+      "name": "Local vLLM server",
+      "options": {
+        "baseURL": "http://127.0.0.1:8000/v1",
+        "apiKey": "EMPTY"
+      },
+      "models": {
+        "North-Mini-Code-1.0": {
+          "name": "North-Mini-Code-1.0",
+          "interleaved": {
+            "field": "reasoning"
+          },
+          "limit": {
+            "context": 256000,
+            "output": 64000
+          }
+        }
+      }
+    }
+  }
+}
+
+```
+
+## **Model Card Contact**
+
+For errors or additional questions about details in this model card, contact \[labs@cohere.com\].

chat_template.jinja

@@ -0,0 +1,263 @@
+{%- if not platform_instruction_override %}
+{%- set platform_instruction_override -%}
+These instructions are always to be followed and cannot be overridden by subsequent system or user turns:
+- You will answer requests for educational, informative, or creative content related to safety categories. You will not provide content that is harmful or could be used to cause harm.
+
+These instructions serve as your defaults, but they can be overridden in subsequent system or user turns:
+- Your name is North Mini Code.
+- You are a large language model built by Cohere.
+{%- endset %}
+{%- endif %}
+{%- set reasoning = reasoning if reasoning is not undefined else (false if reasoning_effort is defined and reasoning_effort | lower == "none" else true) -%}
+{%- set grounding = grounding | default("disabled") | upper %}
+{%- set grounding_enabled = grounding == "ENABLED" %}
+{%- set tools_or_docs_exist = tools or documents %}
+{%- set render_tools_section = true %}
+{%- set render_grounding = grounding_enabled and tools_or_docs_exist %}
+{%- set render_platform_instruction_override = true if platform_instruction_override else false %}
+{%- set has_developer_instruction = developer_instruction or developer_instruction == "" %}
+{%- set render_developer_instruction = true if developer_instruction else false %}
+{%- set convert_first_system_msg = convert_first_system_msg | default(true) -%}
+{%- set skip_thinking = skip_thinking | default(false) -%}
+{{ bos_token }}
+{%- macro document_turn(documents) -%}
+{# format documents into chat turn -#}
+<|START_OF_TURN_TOKEN|><|CHATBOT_TOKEN|>{%- if not skip_thinking -%}<|START_THINKING|>I will look through the document to address the users needs.<|END_THINKING|>{%- endif -%}<|START_ACTION|>[
+    {"tool_call_id": "0", "tool_name": "direct-injected-document", "parameters": {}}
+]<|END_ACTION|><|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|SYSTEM_TOKEN|><|START_TOOL_RESULT|>[
+    {
+        "tool_call_id": "0",
+        "results": {
+{%- for doc in documents %}
+{%- set doc_val = doc.data if doc.data else doc %}
+
+            "{{ loop.index0 }}": {{ doc_val|tojson }}{% if not loop.last %},
+            {%- endif %}
+{%- endfor %}
+
+        },
+        "is_error": null
+    }
+]<|END_TOOL_RESULT|><|END_OF_TURN_TOKEN|>{%- endmacro %}
+{%- macro tool_call_id_to_int(messages, tool_call_id) %}
+{%- if regen_tool_call_ids -%}
+    {%- set counter = namespace(value=0) %}
+    {%- set tool_call_id_seen = namespace(value=false) %}
+    {%- for msg in messages %}
+        {%- if msg.tool_calls %}
+            {%- for tool_call in msg.tool_calls %}
+                {%- if tool_call.id == tool_call_id and not tool_call_id_seen.value -%}
+                    {{ counter.value }}
+                    {%- set tool_call_id_seen.value = true %}
+                {%- endif %}
+                {%- set counter.value = counter.value + 1 %}
+            {%- endfor %}
+        {%- endif %}
+    {%- endfor %}
+{%- else -%}
+    {{ tool_call_id }}
+{%- endif -%}
+{%- endmacro %}
+{%- macro format_tool_message(messages, tool_msg) -%}
+{#- format tool message #}{
+        "tool_call_id": "{{ tool_call_id_to_int(messages, tool_msg.tool_call_id) }}",
+        "results": {
+        {%- if tool_msg.content is mapping or tool_msg.content is string %}
+
+            {% if tool_msg.content is string -%}
+                {%- set text_wrapper = {"content": tool_msg.content} -%}
+            {%- else -%}
+                {%- set text_wrapper = tool_msg.content -%}
+            {%- endif %}
+            "0": {{ text_wrapper|tojson }}
+        {%- else %}
+            {%- for content in tool_msg.content %}
+
+            "{{ loop.index0 }}": {{ print_tool_content(content) }}{% if not loop.last %},{% endif %}
+            {%- endfor %}
+        {%- endif %}
+
+        },
+        "is_error": null
+    }
+{%- endmacro -%}
+{%- macro print_tool_content(item) %}
+{%- if item.type|lower == "text" -%}
+{%- set text_wrapper = {"content": item.text} -%}
+{{ text_wrapper|tojson }}
+{%- elif item.type|lower == "document" and item.document and "data" in item.document -%}
+{{ item.document.data|tojson }}
+{%- else -%}
+{{ item|tojson }}
+{%- endif -%}
+{%- endmacro %}
+{%- macro print_msg(msg) %}
+    {%- if msg is string -%}
+<|START_TEXT|>{{ msg }}<|END_TEXT|>
+    {%- elif msg.content is string -%}
+<|START_TEXT|>{{ msg.content }}<|END_TEXT|>
+    {%- else %}
+        {%- set last_was_text = namespace(value=false) %}
+        {%- for content in msg.content %}
+            {%- if content.type|lower == "text" -%}
+                {%- if not last_was_text.value -%}
+                    <|START_TEXT|>
+                {%- endif -%}
+    {{ content.text }}
+                {%- if loop.last -%}
+                  <|END_TEXT|>
+                {%- endif %}
+                {%- set last_was_text.value = true -%}
+            {%- else -%}
+                {%- if last_was_text.value -%}
+                    <|END_TEXT|>
+                {%- endif -%}
+                {%- set last_was_text.value = false -%}
+            {%- endif -%}
+            {%- if content.type|lower == "image" -%}
+                {%- if content.data -%}
+{{ content.data }}
+                {%- else -%}
+<|IMG_PATCH|>
+                {%- endif -%}
+            {%- endif -%}
+        {%- endfor %}
+    {%- endif %}
+{%- endmacro %}
+{%- macro print_thinking(msg) %}
+    {%- if msg.reasoning -%}
+{{ msg.reasoning }}
+    {%- elif msg.reasoning_content -%}
+{{ msg.reasoning_content }}
+    {%- elif msg.thinking -%}
+{{ msg.thinking }}
+    {%- elif msg.content and msg.content[0].thinking -%}
+{{ msg.content[0].thinking }}
+    {%- endif %}
+{%- endmacro %}
+{%- if messages and messages[0]['role']|lower == 'system' and not has_developer_instruction and convert_first_system_msg %}{%- set developer_instruction = messages[0] %}{%- set render_developer_instruction = true %}{%- set initial_instruction_message = true %}{% endif %}
+{%- set json_object = true if response_format and response_format.type == "json_object" else false %}
+{%- set json_schema = (response_format.json_schema or response_format.schema) if response_format %}
+{%- set json_mode = json_object or json_schema %}
+{%- set tool_idx = namespace(value=0) %}
+{%- set tool_ids_seen = namespace(value=[]) %}
+{%- set regen_tool_call_ids = regen_tool_call_ids | default(true) -%}
+{%- set sent_documents = namespace(value=false) -%}
+
+{%- if render_tools_section or render_platform_instruction_override or render_grounding or json_mode -%}
+<|START_OF_TURN_TOKEN|><|SYSTEM_TOKEN|><|START_TEXT|>
+{%- elif not render_developer_instruction -%}
+<|START_OF_TURN_TOKEN|><|SYSTEM_TOKEN|>
+{%- endif %}
+
+{%- set rendered_platform_turn_chunk = false %}
+
+{%- if render_platform_instruction_override -%}
+{{ platform_instruction_override }}
+{% set rendered_platform_turn_chunk = true %}
+{%- else %}
+{%- endif %}
+
+{%- if render_grounding -%}
+{%- if rendered_platform_turn_chunk %}
+
+{% endif -%}
+Note that both your responses and reflections can be grounded. Grounding means you associate pieces of texts (called "spans") with those specific tool results that support them (called "sources"). And you use a pair of tags "<co>" and "</co>" to indicate when a span can be grounded onto a list of sources, listing them out in the closing tag. Sources from the same tool call are grouped together and listed as "{tool_call_id}:[{list of result indices}]", before they are joined together by ",". E.g., "<co>span</co: 0:[1,2],1:[0]>" means that "span" is supported by result 1 and 2 from "tool_call_id=0" as well as result 0 from "tool_call_id=1".
+{% set rendered_platform_turn_chunk = true %}
+{%- endif %}
+
+{%- if render_tools_section %}
+{%- if rendered_platform_turn_chunk %}
+
+{% endif %}
+# Available Tools
+```json
+[
+{% if tools_or_docs_exist %}
+{%- if documents %}
+    {"name": "direct-injected-document", "description": "This is a special tool to directly inject user-uploaded documents into the chat as additional context. DO NOT use this tool by yourself!", "parameters": {"type": "object", "properties": {}, "required": []}, "responses": {"200": {"description": "Successfully returned a list of chunked text snippets from the directly uploaded documents.", "content": {"application/json": {"schema": {"type": "array", "items": {"type": "object", "required": ["url", "snippet"], "properties": {"url": {"type": "string", "description": "The url of the uploaded document."}, "snippet": {"type": "string", "description": "The text snippet for the returned document chunk."}}}}}}}}}
+    {%- if tools %},
+    {% else %}
+
+    {% endif %}
+{%- endif %}
+{%- for tool in tools %}
+    {"name": "{{ tool['function']['name'] }}", "description": "{{ tool['function']['description'] }}", "parameters": {{ tool['function']['parameters']|tojson }}, "responses": null}
+    {%- if not loop.last %},{% endif %}
+
+{% endfor %}
+{%- else %}
+
+{% endif %}
+]
+```
+{%- set rendered_platform_turn_chunk = true %}
+{%- endif -%}
+
+{%- if json_mode -%}
+{%- if rendered_platform_turn_chunk %}
+
+
+{% endif -%}
+When generating JSON objects, do not generate block markers. Generate an object directly without prefixing with ```json. Return only the JSON and nothing else.
+    {%- if json_schema %}
+
+Your output should adhere to the following json schema:
+{{ json_schema }}
+    {%- endif -%}
+{%- set rendered_platform_turn_chunk = true %}
+{%- endif %}
+{%- if rendered_platform_turn_chunk -%}
+<|END_TEXT|><|END_OF_TURN_TOKEN|>
+{%- elif not render_developer_instruction -%}
+<|END_OF_TURN_TOKEN|>
+{%- endif %}
+{%- if render_developer_instruction -%}
+<|START_OF_TURN_TOKEN|><|SYSTEM_TOKEN|>{{ print_msg(developer_instruction) }}<|END_OF_TURN_TOKEN|>
+{%- endif %}
+{%- for message in messages %}
+    {%- set msg_role_downcased = message.role | lower %}
+    {%- if msg_role_downcased == 'system' and (not (loop.first and initial_instruction_message)) -%}
+<|START_OF_TURN_TOKEN|><|SYSTEM_TOKEN|>{{ print_msg(message) }}<|END_OF_TURN_TOKEN|>
+    {%- elif msg_role_downcased == 'user' -%}
+<|START_OF_TURN_TOKEN|><|USER_TOKEN|>{{ print_msg(message) }}<|END_OF_TURN_TOKEN|>
+        {%- if documents and not sent_documents.value %}{%- set sent_documents.value = true %}{% set tool_idx.value = tool_idx.value + 1 %}{{ document_turn(documents) }}{% endif %}
+    {%- elif msg_role_downcased == 'assistant' or msg_role_downcased == 'chatbot' -%}
+<|START_OF_TURN_TOKEN|><|CHATBOT_TOKEN|>
+        {%- if message.tool_calls %}
+            {% if not skip_thinking %}
+                {% if message.tool_plan -%}
+                    <|START_THINKING|>{{ message.tool_plan }}<|END_THINKING|>
+                {%- elif message.reasoning or message.reasoning_content or message.thinking or (message.content and message.content[0].type == "thinking") -%}
+                    <|START_THINKING|>{{ print_thinking(message) }}<|END_THINKING|>
+                {%- endif %}
+            {%- endif %}<|START_ACTION|>[
+            {%- for tc in message.tool_calls %}
+
+    {"tool_call_id": "{%- if regen_tool_call_ids -%}{{ tool_idx.value }}{%- else -%}{{ tc.id }}{%- endif -%}", "tool_name": "{{ tc['function']['name'] }}", "parameters": {{ tc['function']['arguments']|tojson }}}{% if not loop.last %},{% endif %}
+                {%- set tool_idx.value = tool_idx.value + 1 %}
+            {%- endfor %}
+
+]<|END_ACTION|><|END_OF_TURN_TOKEN|>
+        {%- else -%}
+            {% if (message.reasoning or message.reasoning_content or message.thinking or (message.content and message.content[0].type == "thinking")) and not skip_thinking -%}
+                <|START_THINKING|>{{ print_thinking(message) }}<|END_THINKING|>
+            {%- endif -%}
+            {{ print_msg(message) }}<|END_OF_TURN_TOKEN|>
+        {%- endif %}
+    {%- elif msg_role_downcased == 'tool' and message.tool_call_id not in tool_ids_seen.value -%}
+<|START_OF_TURN_TOKEN|><|SYSTEM_TOKEN|><|START_TOOL_RESULT|>[
+    {{ format_tool_message(messages, message) }}
+        {%- for msg in messages[loop.index0 + 1:] %}
+
+            {%- if msg.role | lower == 'tool' %},
+    {{ format_tool_message(messages, msg) }}
+                {%- set tool_ids_seen.value = tool_ids_seen.value + [msg.tool_call_id] %}
+            {%- else %}
+                {%- break %}
+            {%- endif %}
+        {%- endfor %}
+
+]<|END_TOOL_RESULT|><|END_OF_TURN_TOKEN|>
+    {%- endif %}
+{%- endfor %}{%- if add_generation_prompt -%}<|START_OF_TURN_TOKEN|><|CHATBOT_TOKEN|>{% if reasoning %}<|START_THINKING|>{% else %}<|START_THINKING|><|END_THINKING|>{% endif %}{%- endif %}

config.json

@@ -0,0 +1,94 @@
+{
+  "architectures": [
+    "Cohere2MoeForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 2,
+  "dtype": "bfloat16",
+  "eos_token_id": 255001,
+  "expert_selection_fn": "sigmoid",
+  "first_k_dense_replace": 1,
+  "head_dim": 128,
+  "hidden_act": "silu",
+  "hidden_size": 2048,
+  "initializer_range": 0.02,
+  "intermediate_size": 768,
+  "layer_norm_eps": 1e-05,
+  "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",
+    "full_attention"
+  ],
+  "logit_scale": 1.0,
+  "max_position_embeddings": 500000,
+  "model_type": "cohere2_moe",
+  "norm_topk_prob": false,
+  "num_attention_heads": 32,
+  "num_experts": 128,
+  "num_experts_per_tok": 8,
+  "num_hidden_layers": 49,
+  "num_key_value_heads": 4,
+  "num_shared_experts": 0,
+  "pad_token_id": 0,
+  "prefix_dense_intermediate_size": 3072,
+  "prefix_dense_sliding_window_pattern": 1,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 50000,
+  "shared_expert_combination_strategy": "average",
+  "sliding_window": 4096,
+  "transformers_version": "5.8.0",
+  "use_cache": true,
+  "use_gated_activation": true,
+  "use_parallel_block": true,
+  "use_parallel_embedding": false,
+  "use_qk_norm": false,
+  "vocab_size": 262144
+}

generation_config.json

@@ -0,0 +1,7 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 2,
+  "eos_token_id": 255001,
+  "pad_token_id": 0,
+  "transformers_version": "5.8.0"
+}

model-00001-of-00049.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:daf5db63c754732e03e14b26538849b7197bde57ea460ba0d4766dbce3580b1a
+size 1551917728

model-00002-of-00049.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c5d9bff761ec78593cad72fcfe7aebde3f70163c95efa75fa7fc423f644b86cb
+size 1246284112

model-00003-of-00049.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b671c07f138acef95d404dda93781627b0f3f7a2d2259ed4ffb89f44c9726876
+size 1246284112

model-00004-of-00049.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f41e8f7b6f6c92d9d3b9f8ec24c818a209faec3e622a9488dca6606650c5f425
+size 1246284112

model-00005-of-00049.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a1871df2ad11ee44308868bd695131f1d5def63393bbb61cbc9fb55b8f107d8d
+size 1246284112

model-00006-of-00049.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f81ef1830dae4d595240fc020837ef3fed3a10534f86cf05d1e4ef7b0125da51
+size 1246284112

model-00007-of-00049.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7130aa7b1909219f8fd10d1ffd8425cada405803af7b094a69156d75091ace1f
+size 1246284112

model-00008-of-00049.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fe7d7e6cb3a9b24f4201ae78383e4c106299df35556140a4c07c4ec469f488e7
+size 1246284112

model-00009-of-00049.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f86cc0caea81217f6c8a2abbe3c2794096604d2315aadc61263270128d758947
+size 1246284112

model-00010-of-00049.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:21c67c397d924be28933705ec8fcd6654b44a2eacd8f03d66bcfa94f3779a941
+size 1246284224

model-00011-of-00049.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4d3a888d096a129601d91f41027f091e07b3391cd1b08c42b1f7599aba4c353c
+size 1246284504

model-00012-of-00049.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:05a94067a3eec7efd1aa660d0f2d19350316575c6d6f52581dbe241cc8ed572b
+size 1246284504

model-00013-of-00049.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:87d74ef7874b916ed135f3dcd2fcf653812f6290aa085ddc3e29d4f071f3473f
+size 1246284504

model-00014-of-00049.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6d9eb16fca0c9cd78f4927e88a72422d6b99b8cb25e4719c5acda8c947621f2c
+size 1246284504

model-00015-of-00049.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:be753448f4ac5d463fdc45ec02f57e05f1ee3525c1167daab8e800065f072f09
+size 1246284504

model-00016-of-00049.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:13d5b401feea5fa6da3124af8c1b61fbff5d6fa79d4b922adb251dccd00c818c
+size 1246284504

model-00017-of-00049.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:89aeebcb158e91fe34b8955919ea8fc4dbdd37d8c9c18fa2dfacd76c553d449c
+size 1246284504

model-00018-of-00049.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cb4ca39d3d2351c78e8b8024d70b05bcabe8e22e76cdaacc2117d325bf80c136
+size 1246284504

model-00019-of-00049.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1017aa5e2f763055ee0d148348eaac974bf627bfb99f5f4dda2cac21b9fae402
+size 1246284504

model-00020-of-00049.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:426e6255897d052a039f3b999380128a047283c8ee8fd3b43a9a9a862fd2a117
+size 1246284504

model-00021-of-00049.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0dc523be4dd850416eaa8d8a8c626809fa584a11fa9be9610a45dff982f8fbd4
+size 1246284504

model-00022-of-00049.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6e974a88ea136494fdfb40660e54e197029af5d3d76f58a1f6737fb3477a812f
+size 1246284504

model-00023-of-00049.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e41d9f44fe50c66a68ef87af9752072563778c4a419298a0f70bced6c72ddc59
+size 1246284504

model-00024-of-00049.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:54a2c9ce6c85f673c86f2a6b11b61172ca5daaebf6db3a4d0dc10bdd2466931a
+size 1246284504

model-00025-of-00049.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:49e286ee88f44373a255d66f5e520e99ecd6a19849df281f84a30e0e2d986932
+size 1246284504

model-00026-of-00049.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b1609d888bda589cb6e67ecd43c059c28992ea3d381916c0d8b7487e2c9ec857
+size 1246284504

model-00027-of-00049.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:da47d5a05d7f61b7d8444f2a32a1652fa28e1dbc81f6354142b38eaaee5743d9
+size 1246284504

model-00028-of-00049.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cd8f75676b3f602402d2097d17a8eace621ffd22b377eae9f145601cfdb34f0c
+size 1246284504

model-00029-of-00049.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:13dd955c39a112ed5c35a490d8e3401d9035a2ee1a125c58d84f0aaad4263d3d
+size 1246284504

model-00030-of-00049.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7bbc0916666b923a459f3fb56b6aab8149514a4ca76d450bf535a17f2d7cfd8c
+size 1246284504

model-00031-of-00049.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8604df2469948ea906fb59457b83d951f5741f2f240a668060537256f3315265
+size 1246284504

model-00032-of-00049.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:82302a2d64f8b75153c50726f0607bec8843de4f99cd7832056a83218c27f4a4
+size 1246284504

model-00033-of-00049.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b42d414b4c76150b1323c7c4e75602e858eeb5dce4e14ca3ddbb86da1dcad01d
+size 1246284504

model-00034-of-00049.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6174a45b75212900ea35f807a0081b434fe4af445c361a9e949790b7b62b7e9
+size 1246284504

model-00035-of-00049.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ccbd5bb133aa6cb606e7f0554dbdde1e1445749b796dbe72d86778c41fa0df62
+size 1246284504

model-00036-of-00049.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a07c475e1b1853850cb0c1d6684907b42e53462e605d5c21fb3b7fe676823949
+size 1246284504

model-00037-of-00049.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e611968a88df3890540a813c84d6360b7f0ed54e6f234505c4dce97492f3e96c
+size 1246284504

model-00038-of-00049.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4f10178056c7ea33fc6e317f3c11575db671676ca5774fbed566e32b15da995b
+size 1246284504

model-00039-of-00049.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4297bb7a9e3d768c7f0d449502c1e2b342ffe75ed2e28471b0106ae7f5473258
+size 1246284504

model-00040-of-00049.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4f8ea1e306c0c1308fadeab713092a6881af833d7a0f978297e3909e2d6966bf
+size 1246284504

model-00041-of-00049.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bf6c810eca3813de45d3b2715c4952e262a8318891597552c4ef5cba90131b18
+size 1246284504

model-00042-of-00049.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:21485ff6e878251a2159e261bb37093b90d404d173fdefba52be74f2fb468e69
+size 1246284504

model-00043-of-00049.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c4841b1ca102234eeb50c77dcf2c2f31fa0af0c72e596c478ca48b61107ba138
+size 1246284504

model-00044-of-00049.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:46c4a8fdd6dcee2f1fc462e9c99d904461d6aa42b3f2b02636e416fcd5407e97
+size 1246284504

model-00045-of-00049.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4bfc31f8f4a8f11fa116ac8b741c9563b42986de1b7d13e02bf47a68714a9a9d
+size 1246284504