gemma4-e4b-toolcall-v01 API & Inference Endpoint

🎯 What This Model Does

This model generates structured tool calls in a compact format when given a user query and available tool definitions.

Output format:

text

call:function_name{param1:value1,param2:value2}

Example:

text

Input: "What's the weather in Tokyo?"Output: call:get_weather{city:Tokyo}

📊 Evaluation Results (v0.1)

Evaluated on held-out validation set (200 samples):

Table with columns: Metric, Score
Metric	Score
Tool Selection Accuracy	64.2%
Full Match (name + args)	28.4%
No-Call Accuracy (avoids hallucination)	69.9%
Missed Tool Call Rate	35.8%

Strengths:

✅ Learned when NOT to call tools (70% no-call accuracy, low hallucination)
✅ Generates structured tool calls (not free-form text)
✅ Selects correct tool ~64% of the time from multiple options

Known Limitations:

⚠️ Uses compact format (call:name{args}) rather than standard JSON
⚠️ Misses tool calls ~36% of the time (responds with text instead)
⚠️ Argument extraction needs improvement (28% full match)
⚠️ v0.1 — not production-ready, experimental release

🚀 Quick Start

Option 1: Merged Model (recommended)

python

from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfigimport torch, json # Load merged model (no adapter needed)bnb = BitsAndBytesConfig(load_in_4bit=True, bnb_4bit_quant_type="nf4", bnb_4bit_compute_dtype=torch.float16, bnb_4bit_use_double_quant=True)model = AutoModelForCausalLM.from_pretrained("roshangrewal/gemma4-e4b-toolcall-v01", quantization_config=bnb, device_map="auto", torch_dtype=torch.float16)tokenizer = AutoTokenizer.from_pretrained("roshangrewal/gemma4-e4b-toolcall-v01") # Define toolstools = [{"type": "function", "function": {"name": "get_weather", "description": "Get weather for a city", "parameters": {"type": "object", "properties": {"city": {"type": "string"}}, "required": ["city"]}}}] messages = [    {"role": "system", "content": f"You have access to these tools:\n{json.dumps(tools)}\nCall the appropriate function when needed."},    {"role": "user", "content": "What's the weather in Mumbai?"}] text = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)inputs = tokenizer(text, return_tensors="pt").to(model.device)with torch.no_grad():    out = model.generate(**inputs, max_new_tokens=200, do_sample=False, pad_token_id=tokenizer.pad_token_id)print(tokenizer.decode(out[0][inputs["input_ids"].shape[1]:], skip_special_tokens=True))# Output: call:get_weather{city:Mumbai}

Option 2: Adapter (lightweight, 51MB)

python

from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfigfrom peft import PeftModelimport torch, json # Load base + adapterbnb = BitsAndBytesConfig(load_in_4bit=True, bnb_4bit_quant_type="nf4", bnb_4bit_compute_dtype=torch.float16, bnb_4bit_use_double_quant=True)base = AutoModelForCausalLM.from_pretrained("google/gemma-4-E4B-it", quantization_config=bnb, device_map="auto", torch_dtype=torch.float16)tokenizer = AutoTokenizer.from_pretrained("google/gemma-4-E4B-it")model = PeftModel.from_pretrained(base, "roshangrewal/gemma4-e4b-toolcall-v01-lora")model.eval()

Deployment Options

Table with columns: Method, VRAM Required, Speed
Method	VRAM Required	Speed
4-bit quantized (above)	~10 GB	Good for T4/4090
fp16 (full precision)	~16 GB	Best quality, needs A10+
GGUF via llama.cpp/Ollama	~6 GB	CPU + GPU hybrid

💬 Prompt Examples

Single tool, simple query

python

tools = [    {"type": "function", "function": {"name": "get_weather", "description": "Get current weather",     "parameters": {"type": "object", "properties": {"city": {"type": "string"}, "unit": {"type": "string", "enum": ["celsius", "fahrenheit"]}}, "required": ["city"]}}}]messages = [    {"role": "system", "content": f"You have access to these tools:\n{json.dumps(tools)}\nCall the appropriate function when needed. When no tool is needed, respond directly."},    {"role": "user", "content": "What's the weather in Tokyo?"}]# Output: call:get_weather{city:Tokyo}

Multiple tools — model selects the right one

python

tools = [    {"type": "function", "function": {"name": "get_weather", "description": "Get weather for a city",     "parameters": {"type": "object", "properties": {"city": {"type": "string"}}, "required": ["city"]}}},    {"type": "function", "function": {"name": "search_web", "description": "Search the web",     "parameters": {"type": "object", "properties": {"query": {"type": "string"}}, "required": ["query"]}}},    {"type": "function", "function": {"name": "send_email", "description": "Send an email",     "parameters": {"type": "object", "properties": {"to": {"type": "string"}, "subject": {"type": "string"}, "body": {"type": "string"}}, "required": ["to", "subject", "body"]}}}]messages = [    {"role": "system", "content": f"You have access to these tools:\n{json.dumps(tools)}\nCall the appropriate function when needed. When no tool is needed, respond directly."},    {"role": "user", "content": "Search for latest news about AI startups in India"}]# Output: call:search_web{query:latest news AI startups India}

No tool needed — model responds directly

python

messages = [    {"role": "system", "content": f"You have access to these tools:\n{json.dumps(tools)}\nCall the appropriate function when needed. When no tool is needed, respond directly."},    {"role": "user", "content": "What is 2 + 2?"}]# Output: 4 (no tool call generated)

Prompt structure (for custom integrations)

text

System: You have access to these tools:[tool definitions as JSON array]Call the appropriate function when needed. When no tool is needed, respond directly. User: <query> | Parameter | Value ||-----------|-------|| Base Model | google/gemma-4-E4B-it (8B params, 4.5B effective) || Method | QLoRA (4-bit NF4, double quantization) || LoRA Rank | 16 || LoRA Alpha | 16 || Target Modules | q_proj.linear, k_proj.linear, v_proj.linear, o_proj.linear, up_proj.linear, down_proj.linear || Learning Rate | 1e-4 (linear decay, 10% warmup) || Effective Batch Size | 16 || Max Length | 1024 || Steps | 10,000 (~84% of 1 epoch) || Training Time | 56 hours || GPU | NVIDIA Tesla T4 (16GB) || Cost | ~$0 (own hardware) | ## 📚 Training Data 174,853 function-calling examples from: | Dataset | Examples ||---------|----------|| [Glaive Function Calling v2](https://huggingface.co/datasets/glaiveai/glaive-function-calling-v2) | 112,960 || [Salesforce xLAM 60K](https://huggingface.co/datasets/Salesforce/xlam-function-calling-60k) | 60,000 || [Hermes Function Calling v1](https://huggingface.co/datasets/NousResearch/hermes-function-calling-v1) | 1,893 | ## 🗺️ Roadmap - **v0.1** (current): Initial fine-tune, compact call format- **v0.2** (planned): Align with Gemma 4's native tool-calling template, target 80%+ accuracy- **v1.0** (planned): Production-ready with BFCL leaderboard submission ## 💡 Parsing the Output ```pythonimport re def parse_tool_call(text):    m = re.findall(r'call:(\w+)\{(.+?)\}', text)    if m:        name = m[0][0]        args = dict(re.findall(r'(\w+):([^,}]+)', m[0][1]))        return {"name": name, "arguments": args}    return None result = parse_tool_call("call:get_weather{city:Tokyo}")# {'name': 'get_weather', 'arguments': {'city': 'Tokyo'}}

🔗 Links

Training Code: github.com/roshangrewal/f-for-finetuning
Base Model: google/gemma-4-E4B-it

📝 Citation

bibtex

@misc{grewal2026gemma4toolcall,  title={Gemma 4 E4B Tool-Calling Fine-Tune v0.1},  author={Roshan Grewal},  year={2026},  url={https://huggingface.co/roshangrewal/gemma4-e4b-toolcall-v01}}

text

**Tips:**- Always include tool definitions in the system message as a JSON array- The system message must contain the instruction "Call the appropriate function when needed"- Model outputs `call:function_name{param:value}` format when it decides to use a tool- Model responds with plain text when no tool is appropriate ## 🏗️ Training Details | Parameter | Value ||-----------|-------|| Base Model | google/gemma-4-E4B-it (8B params, 4.5B effective) || Method | QLoRA (4-bit NF4, double quantization) || LoRA Rank | 16 || LoRA Alpha | 16 || Target Modules | q_proj.linear, k_proj.linear, v_proj.linear, o_proj.linear, up_proj.linear, down_proj.linear || Learning Rate | 1e-4 (linear decay, 10% warmup) || Effective Batch Size | 16 || Max Length | 1024 || Steps | 10,000 (~84% of 1 epoch) || Training Time | 56 hours || GPU | NVIDIA Tesla T4 (16GB) | ## 📚 Training Data 174,853 function-calling examples from: | Dataset | Examples ||---------|----------|| [Glaive Function Calling v2](https://huggingface.co/datasets/glaiveai/glaive-function-calling-v2) | 112,960 || [Salesforce xLAM 60K](https://huggingface.co/datasets/Salesforce/xlam-function-calling-60k) | 60,000 || [Hermes Function Calling v1](https://huggingface.co/datasets/NousResearch/hermes-function-calling-v1) | 1,893 | ## 🗺️ Roadmap - **v0.1** (current): Initial fine-tune, compact call format- **v0.2** (planned): Align with Gemma 4's native tool-calling template, target 80%+ accuracy- **v1.0** (planned): Production-ready with BFCL leaderboard submission ## 💡 Parsing the Output ```pythonimport re def parse_tool_call(text):    m = re.findall(r'call:(\w+)\{(.+?)\}', text)    if m:        name = m[0][0]        args = dict(re.findall(r'(\w+):([^,}]+)', m[0][1]))        return {"name": name, "arguments": args}    return None result = parse_tool_call("call:get_weather{city:Tokyo}")# {'name': 'get_weather', 'arguments': {'city': 'Tokyo'}}

🔗 Links

Training Code: github.com/roshangrewal/f-for-finetuning
Base Model: google/gemma-4-E4B-it

📝 Citation

bibtex

@misc{grewal2026gemma4toolcall,  title={Gemma 4 E4B Tool-Calling Fine-Tune v0.1},  author={Roshan Grewal},  year={2026},  url={https://huggingface.co/roshangrewal/gemma4-e4b-toolcall-v01}}

🎯 What This Model Does

This model generates structured tool calls in a compact format when given a user query and available tool definitions.

Output format:

text

call:function_name{param1:value1,param2:value2}

Example:

text

Input: "What's the weather in Tokyo?"Output: call:get_weather{city:Tokyo}

📊 Evaluation Results (v0.1)

Evaluated on held-out validation set (200 samples):

Table with columns: Metric, Score
Metric	Score
Tool Selection Accuracy	64.2%
Full Match (name + args)	28.4%
No-Call Accuracy (avoids hallucination)	69.9%
Missed Tool Call Rate	35.8%

Strengths:

✅ Learned when NOT to call tools (70% no-call accuracy, low hallucination)
✅ Generates structured tool calls (not free-form text)
✅ Selects correct tool ~64% of the time from multiple options

Known Limitations:

⚠️ Uses compact format (call:name{args}) rather than standard JSON
⚠️ Misses tool calls ~36% of the time (responds with text instead)
⚠️ Argument extraction needs improvement (28% full match)
⚠️ v0.1 — not production-ready, experimental release

🚀 Quick Start

Option 1: Merged Model (recommended)

python

from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfigimport torch, json # Load merged model (no adapter needed)bnb = BitsAndBytesConfig(load_in_4bit=True, bnb_4bit_quant_type="nf4", bnb_4bit_compute_dtype=torch.float16, bnb_4bit_use_double_quant=True)model = AutoModelForCausalLM.from_pretrained("roshangrewal/gemma4-e4b-toolcall-v01", quantization_config=bnb, device_map="auto", torch_dtype=torch.float16)tokenizer = AutoTokenizer.from_pretrained("roshangrewal/gemma4-e4b-toolcall-v01") # Define toolstools = [{"type": "function", "function": {"name": "get_weather", "description": "Get weather for a city", "parameters": {"type": "object", "properties": {"city": {"type": "string"}}, "required": ["city"]}}}] messages = [    {"role": "system", "content": f"You have access to these tools:\n{json.dumps(tools)}\nCall the appropriate function when needed."},    {"role": "user", "content": "What's the weather in Mumbai?"}] text = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)inputs = tokenizer(text, return_tensors="pt").to(model.device)with torch.no_grad():    out = model.generate(**inputs, max_new_tokens=200, do_sample=False, pad_token_id=tokenizer.pad_token_id)print(tokenizer.decode(out[0][inputs["input_ids"].shape[1]:], skip_special_tokens=True))# Output: call:get_weather{city:Mumbai}

from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfigimport torch, json # Load merged model (no adapter needed)bnb = BitsAndBytesConfig(load_in_4bit=True, bnb_4bit_quant_type="nf4", bnb_4bit_compute_dtype=torch.float16, bnb_4bit_use_double_quant=True)model = AutoModelForCausalLM.from_pretrained("roshangrewal/gemma4-e4b-toolcall-v01", quantization_config=bnb, device_map="auto", torch_dtype=torch.float16)tokenizer = AutoTokenizer.from_pretrained("roshangrewal/gemma4-e4b-toolcall-v01") # Define toolstools = [{"type": "function", "function": {"name": "get_weather", "description": "Get weather for a city", "parameters": {"type": "object", "properties": {"city": {"type": "string"}}, "required": ["city"]}}}] messages = [    {"role": "system", "content": f"You have access to these tools:\n{json.dumps(tools)}\nCall the appropriate function when needed."},    {"role": "user", "content": "What's the weather in Mumbai?"}] text = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)inputs = tokenizer(text, return_tensors="pt").to(model.device)with torch.no_grad():    out = model.generate(**inputs, max_new_tokens=200, do_sample=False, pad_token_id=tokenizer.pad_token_id)print(tokenizer.decode(out[0][inputs["input_ids"].shape[1]:], skip_special_tokens=True))# Output: call:get_weather{city:Mumbai}

Option 2: Adapter (lightweight, 51MB)

python

from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfigfrom peft import PeftModelimport torch, json # Load base + adapterbnb = BitsAndBytesConfig(load_in_4bit=True, bnb_4bit_quant_type="nf4", bnb_4bit_compute_dtype=torch.float16, bnb_4bit_use_double_quant=True)base = AutoModelForCausalLM.from_pretrained("google/gemma-4-E4B-it", quantization_config=bnb, device_map="auto", torch_dtype=torch.float16)tokenizer = AutoTokenizer.from_pretrained("google/gemma-4-E4B-it")model = PeftModel.from_pretrained(base, "roshangrewal/gemma4-e4b-toolcall-v01-lora")model.eval()

from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfigfrom peft import PeftModelimport torch, json # Load base + adapterbnb = BitsAndBytesConfig(load_in_4bit=True, bnb_4bit_quant_type="nf4", bnb_4bit_compute_dtype=torch.float16, bnb_4bit_use_double_quant=True)base = AutoModelForCausalLM.from_pretrained("google/gemma-4-E4B-it", quantization_config=bnb, device_map="auto", torch_dtype=torch.float16)tokenizer = AutoTokenizer.from_pretrained("google/gemma-4-E4B-it")model = PeftModel.from_pretrained(base, "roshangrewal/gemma4-e4b-toolcall-v01-lora")model.eval()

Deployment Options

Table with columns: Method, VRAM Required, Speed
Method	VRAM Required	Speed
4-bit quantized (above)	~10 GB	Good for T4/4090
fp16 (full precision)	~16 GB	Best quality, needs A10+
GGUF via llama.cpp/Ollama	~6 GB	CPU + GPU hybrid

💬 Prompt Examples

Single tool, simple query

python

tools = [    {"type": "function", "function": {"name": "get_weather", "description": "Get current weather",     "parameters": {"type": "object", "properties": {"city": {"type": "string"}, "unit": {"type": "string", "enum": ["celsius", "fahrenheit"]}}, "required": ["city"]}}}]messages = [    {"role": "system", "content": f"You have access to these tools:\n{json.dumps(tools)}\nCall the appropriate function when needed. When no tool is needed, respond directly."},    {"role": "user", "content": "What's the weather in Tokyo?"}]# Output: call:get_weather{city:Tokyo}

tools = [    {"type": "function", "function": {"name": "get_weather", "description": "Get current weather",     "parameters": {"type": "object", "properties": {"city": {"type": "string"}, "unit": {"type": "string", "enum": ["celsius", "fahrenheit"]}}, "required": ["city"]}}}]messages = [    {"role": "system", "content": f"You have access to these tools:\n{json.dumps(tools)}\nCall the appropriate function when needed. When no tool is needed, respond directly."},    {"role": "user", "content": "What's the weather in Tokyo?"}]# Output: call:get_weather{city:Tokyo}

Multiple tools — model selects the right one

python

tools = [    {"type": "function", "function": {"name": "get_weather", "description": "Get weather for a city",     "parameters": {"type": "object", "properties": {"city": {"type": "string"}}, "required": ["city"]}}},    {"type": "function", "function": {"name": "search_web", "description": "Search the web",     "parameters": {"type": "object", "properties": {"query": {"type": "string"}}, "required": ["query"]}}},    {"type": "function", "function": {"name": "send_email", "description": "Send an email",     "parameters": {"type": "object", "properties": {"to": {"type": "string"}, "subject": {"type": "string"}, "body": {"type": "string"}}, "required": ["to", "subject", "body"]}}}]messages = [    {"role": "system", "content": f"You have access to these tools:\n{json.dumps(tools)}\nCall the appropriate function when needed. When no tool is needed, respond directly."},    {"role": "user", "content": "Search for latest news about AI startups in India"}]# Output: call:search_web{query:latest news AI startups India}

tools = [    {"type": "function", "function": {"name": "get_weather", "description": "Get weather for a city",     "parameters": {"type": "object", "properties": {"city": {"type": "string"}}, "required": ["city"]}}},    {"type": "function", "function": {"name": "search_web", "description": "Search the web",     "parameters": {"type": "object", "properties": {"query": {"type": "string"}}, "required": ["query"]}}},    {"type": "function", "function": {"name": "send_email", "description": "Send an email",     "parameters": {"type": "object", "properties": {"to": {"type": "string"}, "subject": {"type": "string"}, "body": {"type": "string"}}, "required": ["to", "subject", "body"]}}}]messages = [    {"role": "system", "content": f"You have access to these tools:\n{json.dumps(tools)}\nCall the appropriate function when needed. When no tool is needed, respond directly."},    {"role": "user", "content": "Search for latest news about AI startups in India"}]# Output: call:search_web{query:latest news AI startups India}

No tool needed — model responds directly

python

messages = [    {"role": "system", "content": f"You have access to these tools:\n{json.dumps(tools)}\nCall the appropriate function when needed. When no tool is needed, respond directly."},    {"role": "user", "content": "What is 2 + 2?"}]# Output: 4 (no tool call generated)

messages = [    {"role": "system", "content": f"You have access to these tools:\n{json.dumps(tools)}\nCall the appropriate function when needed. When no tool is needed, respond directly."},    {"role": "user", "content": "What is 2 + 2?"}]# Output: 4 (no tool call generated)

Prompt structure (for custom integrations)

text

System: You have access to these tools:[tool definitions as JSON array]Call the appropriate function when needed. When no tool is needed, respond directly.User: <query>| Parameter | Value ||-----------|-------|| Base Model | google/gemma-4-E4B-it (8B params, 4.5B effective) || Method | QLoRA (4-bit NF4, double quantization) || LoRA Rank | 16 || LoRA Alpha | 16 || Target Modules | q_proj.linear, k_proj.linear, v_proj.linear, o_proj.linear, up_proj.linear, down_proj.linear || Learning Rate | 1e-4 (linear decay, 10% warmup) || Effective Batch Size | 16 || Max Length | 1024 || Steps | 10,000 (~84% of 1 epoch) || Training Time | 56 hours || GPU | NVIDIA Tesla T4 (16GB) || Cost | ~$0 (own hardware) |## 📚 Training Data174,853 function-calling examples from:| Dataset | Examples ||---------|----------|| [Glaive Function Calling v2](https://huggingface.co/datasets/glaiveai/glaive-function-calling-v2) | 112,960 || [Salesforce xLAM 60K](https://huggingface.co/datasets/Salesforce/xlam-function-calling-60k) | 60,000 || [Hermes Function Calling v1](https://huggingface.co/datasets/NousResearch/hermes-function-calling-v1) | 1,893 |## 🗺️ Roadmap- **v0.1** (current): Initial fine-tune, compact call format- **v0.2** (planned): Align with Gemma 4's native tool-calling template, target 80%+ accuracy- **v1.0** (planned): Production-ready with BFCL leaderboard submission## 💡 Parsing the Output```pythonimport redef parse_tool_call(text):    m = re.findall(r'call:(\w+)\{(.+?)\}', text)    if m:        name = m[0][0]        args = dict(re.findall(r'(\w+):([^,}]+)', m[0][1]))        return {"name": name, "arguments": args}    return Noneresult = parse_tool_call("call:get_weather{city:Tokyo}")# {'name': 'get_weather', 'arguments': {'city': 'Tokyo'}}

System: You have access to these tools:[tool definitions as JSON array]Call the appropriate function when needed. When no tool is needed, respond directly. User: <query> | Parameter | Value ||-----------|-------|| Base Model | google/gemma-4-E4B-it (8B params, 4.5B effective) || Method | QLoRA (4-bit NF4, double quantization) || LoRA Rank | 16 || LoRA Alpha | 16 || Target Modules | q_proj.linear, k_proj.linear, v_proj.linear, o_proj.linear, up_proj.linear, down_proj.linear || Learning Rate | 1e-4 (linear decay, 10% warmup) || Effective Batch Size | 16 || Max Length | 1024 || Steps | 10,000 (~84% of 1 epoch) || Training Time | 56 hours || GPU | NVIDIA Tesla T4 (16GB) || Cost | ~$0 (own hardware) | ## 📚 Training Data 174,853 function-calling examples from: | Dataset | Examples ||---------|----------|| [Glaive Function Calling v2](https://huggingface.co/datasets/glaiveai/glaive-function-calling-v2) | 112,960 || [Salesforce xLAM 60K](https://huggingface.co/datasets/Salesforce/xlam-function-calling-60k) | 60,000 || [Hermes Function Calling v1](https://huggingface.co/datasets/NousResearch/hermes-function-calling-v1) | 1,893 | ## 🗺️ Roadmap - **v0.1** (current): Initial fine-tune, compact call format- **v0.2** (planned): Align with Gemma 4's native tool-calling template, target 80%+ accuracy- **v1.0** (planned): Production-ready with BFCL leaderboard submission ## 💡 Parsing the Output ```pythonimport re def parse_tool_call(text):    m = re.findall(r'call:(\w+)\{(.+?)\}', text)    if m:        name = m[0][0]        args = dict(re.findall(r'(\w+):([^,}]+)', m[0][1]))        return {"name": name, "arguments": args}    return None result = parse_tool_call("call:get_weather{city:Tokyo}")# {'name': 'get_weather', 'arguments': {'city': 'Tokyo'}}

🔗 Links

Training Code: github.com/roshangrewal/f-for-finetuning
Base Model: google/gemma-4-E4B-it

📝 Citation

bibtex

@misc{grewal2026gemma4toolcall,  title={Gemma 4 E4B Tool-Calling Fine-Tune v0.1},  author={Roshan Grewal},  year={2026},  url={https://huggingface.co/roshangrewal/gemma4-e4b-toolcall-v01}}

@misc{grewal2026gemma4toolcall,  title={Gemma 4 E4B Tool-Calling Fine-Tune v0.1},  author={Roshan Grewal},  year={2026},  url={https://huggingface.co/roshangrewal/gemma4-e4b-toolcall-v01}}

text

**Tips:**- Always include tool definitions in the system message as a JSON array- The system message must contain the instruction "Call the appropriate function when needed"- Model outputs `call:function_name{param:value}` format when it decides to use a tool- Model responds with plain text when no tool is appropriate## 🏗️ Training Details| Parameter | Value ||-----------|-------|| Base Model | google/gemma-4-E4B-it (8B params, 4.5B effective) || Method | QLoRA (4-bit NF4, double quantization) || LoRA Rank | 16 || LoRA Alpha | 16 || Target Modules | q_proj.linear, k_proj.linear, v_proj.linear, o_proj.linear, up_proj.linear, down_proj.linear || Learning Rate | 1e-4 (linear decay, 10% warmup) || Effective Batch Size | 16 || Max Length | 1024 || Steps | 10,000 (~84% of 1 epoch) || Training Time | 56 hours || GPU | NVIDIA Tesla T4 (16GB) |## 📚 Training Data174,853 function-calling examples from:| Dataset | Examples ||---------|----------|| [Glaive Function Calling v2](https://huggingface.co/datasets/glaiveai/glaive-function-calling-v2) | 112,960 || [Salesforce xLAM 60K](https://huggingface.co/datasets/Salesforce/xlam-function-calling-60k) | 60,000 || [Hermes Function Calling v1](https://huggingface.co/datasets/NousResearch/hermes-function-calling-v1) | 1,893 |## 🗺️ Roadmap- **v0.1** (current): Initial fine-tune, compact call format- **v0.2** (planned): Align with Gemma 4's native tool-calling template, target 80%+ accuracy- **v1.0** (planned): Production-ready with BFCL leaderboard submission## 💡 Parsing the Output```pythonimport redef parse_tool_call(text):    m = re.findall(r'call:(\w+)\{(.+?)\}', text)    if m:        name = m[0][0]        args = dict(re.findall(r'(\w+):([^,}]+)', m[0][1]))        return {"name": name, "arguments": args}    return Noneresult = parse_tool_call("call:get_weather{city:Tokyo}")# {'name': 'get_weather', 'arguments': {'city': 'Tokyo'}}

**Tips:**- Always include tool definitions in the system message as a JSON array- The system message must contain the instruction "Call the appropriate function when needed"- Model outputs `call:function_name{param:value}` format when it decides to use a tool- Model responds with plain text when no tool is appropriate ## 🏗️ Training Details | Parameter | Value ||-----------|-------|| Base Model | google/gemma-4-E4B-it (8B params, 4.5B effective) || Method | QLoRA (4-bit NF4, double quantization) || LoRA Rank | 16 || LoRA Alpha | 16 || Target Modules | q_proj.linear, k_proj.linear, v_proj.linear, o_proj.linear, up_proj.linear, down_proj.linear || Learning Rate | 1e-4 (linear decay, 10% warmup) || Effective Batch Size | 16 || Max Length | 1024 || Steps | 10,000 (~84% of 1 epoch) || Training Time | 56 hours || GPU | NVIDIA Tesla T4 (16GB) | ## 📚 Training Data 174,853 function-calling examples from: | Dataset | Examples ||---------|----------|| [Glaive Function Calling v2](https://huggingface.co/datasets/glaiveai/glaive-function-calling-v2) | 112,960 || [Salesforce xLAM 60K](https://huggingface.co/datasets/Salesforce/xlam-function-calling-60k) | 60,000 || [Hermes Function Calling v1](https://huggingface.co/datasets/NousResearch/hermes-function-calling-v1) | 1,893 | ## 🗺️ Roadmap - **v0.1** (current): Initial fine-tune, compact call format- **v0.2** (planned): Align with Gemma 4's native tool-calling template, target 80%+ accuracy- **v1.0** (planned): Production-ready with BFCL leaderboard submission ## 💡 Parsing the Output ```pythonimport re def parse_tool_call(text):    m = re.findall(r'call:(\w+)\{(.+?)\}', text)    if m:        name = m[0][0]        args = dict(re.findall(r'(\w+):([^,}]+)', m[0][1]))        return {"name": name, "arguments": args}    return None result = parse_tool_call("call:get_weather{city:Tokyo}")# {'name': 'get_weather', 'arguments': {'city': 'Tokyo'}}

🔗 Links

Training Code: github.com/roshangrewal/f-for-finetuning
Base Model: google/gemma-4-E4B-it

📝 Citation

bibtex

@misc{grewal2026gemma4toolcall,  title={Gemma 4 E4B Tool-Calling Fine-Tune v0.1},  author={Roshan Grewal},  year={2026},  url={https://huggingface.co/roshangrewal/gemma4-e4b-toolcall-v01}}

@misc{grewal2026gemma4toolcall,  title={Gemma 4 E4B Tool-Calling Fine-Tune v0.1},  author={Roshan Grewal},  year={2026},  url={https://huggingface.co/roshangrewal/gemma4-e4b-toolcall-v01}}

gemma4-e4b-toolcall-v01

README

🎯 What This Model Does

📊 Evaluation Results (v0.1)

🚀 Quick Start

Option 1: Merged Model (recommended)

Option 2: Adapter (lightweight, 51MB)

Deployment Options

💬 Prompt Examples

Single tool, simple query

Multiple tools — model selects the right one

No tool needed — model responds directly

Prompt structure (for custom integrations)

🔗 Links

📝 Citation

🔗 Links

📝 Citation

Explore FriendliAI today

README

🎯 What This Model Does

📊 Evaluation Results (v0.1)

🚀 Quick Start

Option 1: Merged Model (recommended)

Option 2: Adapter (lightweight, 51MB)

Deployment Options

💬 Prompt Examples

Single tool, simple query

Multiple tools — model selects the right one

No tool needed — model responds directly

Prompt structure (for custom integrations)

🔗 Links

📝 Citation

🔗 Links

📝 Citation