Text Generation
Transformers
Safetensors
granitemoehybrid
language
granite-4.0
conversational
Eval Results
Instructions to use ibm-granite/granite-4.0-h-tiny with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use ibm-granite/granite-4.0-h-tiny with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="ibm-granite/granite-4.0-h-tiny") messages = [ {"role": "user", "content": "Who are you?"}, ] pipe(messages)# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("ibm-granite/granite-4.0-h-tiny") model = AutoModelForCausalLM.from_pretrained("ibm-granite/granite-4.0-h-tiny") messages = [ {"role": "user", "content": "Who are you?"}, ] inputs = tokenizer.apply_chat_template( messages, add_generation_prompt=True, tokenize=True, return_dict=True, return_tensors="pt", ).to(model.device) outputs = model.generate(**inputs, max_new_tokens=40) print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:])) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use ibm-granite/granite-4.0-h-tiny with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "ibm-granite/granite-4.0-h-tiny" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "ibm-granite/granite-4.0-h-tiny", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/ibm-granite/granite-4.0-h-tiny
- SGLang
How to use ibm-granite/granite-4.0-h-tiny with SGLang:
Install from pip and serve model
# Install SGLang from pip: pip install sglang # Start the SGLang server: python3 -m sglang.launch_server \ --model-path "ibm-granite/granite-4.0-h-tiny" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "ibm-granite/granite-4.0-h-tiny", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker images
docker run --gpus all \ --shm-size 32g \ -p 30000:30000 \ -v ~/.cache/huggingface:/root/.cache/huggingface \ --env "HF_TOKEN=<secret>" \ --ipc=host \ lmsysorg/sglang:latest \ python3 -m sglang.launch_server \ --model-path "ibm-granite/granite-4.0-h-tiny" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "ibm-granite/granite-4.0-h-tiny", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }' - Docker Model Runner
How to use ibm-granite/granite-4.0-h-tiny with Docker Model Runner:
docker model run hf.co/ibm-granite/granite-4.0-h-tiny
model card update
Browse files
README.md
CHANGED
|
@@ -33,6 +33,7 @@ The model is designed to respond to general instructions and can be used to buil
|
|
| 33 |
* Code related tasks
|
| 34 |
* Function-calling tasks
|
| 35 |
* Multilingual dialog use cases
|
|
|
|
| 36 |
|
| 37 |
<!-- <todo>Need to test the examples. (especially the tool calling and RAG ones)</todo>
|
| 38 |
-->
|
|
@@ -149,140 +150,302 @@ For each tool call, return a json object with function name and arguments within
|
|
| 149 |
*Coming soon* -->
|
| 150 |
|
| 151 |
**Evaluation Results:**
|
| 152 |
-
*Coming soon*
|
| 153 |
|
| 154 |
-
<
|
| 155 |
-
<caption><b>
|
| 156 |
<thead>
|
| 157 |
<tr>
|
| 158 |
-
<th style="text-align:left; background-color: #001d6c; color: white;">
|
| 159 |
-
<th style="text-align:
|
| 160 |
-
<th style="text-align:center; background-color: #001d6c; color: white;">
|
| 161 |
-
<th style="text-align:center; background-color: #001d6c; color: white;">
|
| 162 |
-
<th style="text-align:center; background-color: #001d6c; color: white;">
|
| 163 |
-
<th style="text-align:center; background-color: #001d6c; color: white;">
|
| 164 |
-
<th style="text-align:center; background-color: #001d6c; color: white;">GSM8K</th>
|
| 165 |
-
<th style="text-align:center; background-color: #001d6c; color: white;">DROP</th>
|
| 166 |
-
<th style="text-align:center; background-color: #001d6c; color: white;">NQ</th>
|
| 167 |
-
<th style="text-align:center; background-color: #001d6c; color: white;">AGIEval</th>
|
| 168 |
-
<th style="text-align:center; background-color: #001d6c; color: white;">TriviaQA</th>
|
| 169 |
-
<th style="text-align:center; background-color: #001d6c; color: white;">Avg</th>
|
| 170 |
-
</tr></thead>
|
| 171 |
-
<tbody>
|
| 172 |
-
<tr>
|
| 173 |
-
<td style="text-align:left; background-color: #DAE8FF; color: #2D2D2D;">Granite-4.0-H-Tiny</td>
|
| 174 |
-
<td style="text-align:center; background-color: #DAE8FF; color: #2D2D2D;"></td>
|
| 175 |
-
<td style="text-align:center; background-color: #DAE8FF; color: #2D2D2D;"></td>
|
| 176 |
-
<td style="text-align:center; background-color: #DAE8FF; color: #2D2D2D;"></td>
|
| 177 |
-
<td style="text-align:center; background-color: #DAE8FF; color: #2D2D2D;"></td>
|
| 178 |
-
<td style="text-align:center; background-color: #DAE8FF; color: #2D2D2D;"></td>
|
| 179 |
-
<td style="text-align:center; background-color: #DAE8FF; color: #2D2D2D;"></td>
|
| 180 |
-
<td style="text-align:center; background-color: #DAE8FF; color: #2D2D2D;"></td>
|
| 181 |
-
<td style="text-align:center; background-color: #DAE8FF; color: #2D2D2D;"></td>
|
| 182 |
-
<td style="text-align:center; background-color: #DAE8FF; color: #2D2D2D;"></td>
|
| 183 |
-
<td style="text-align:center; background-color: #DAE8FF; color: #2D2D2D;"></td>
|
| 184 |
-
<td style="text-align:center; background-color: #DAE8FF; color: #2D2D2D;"></td>
|
| 185 |
-
</tr>
|
| 186 |
-
<tr>
|
| 187 |
-
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">Granite-4.0-Hybrid-Micro</td>
|
| 188 |
-
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 189 |
-
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 190 |
-
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 191 |
-
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 192 |
-
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 193 |
-
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 194 |
-
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 195 |
-
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 196 |
-
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 197 |
-
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 198 |
-
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 199 |
-
</tr>
|
| 200 |
-
<tr>
|
| 201 |
-
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">Granite-4.0-Hybrid-Tiny</td>
|
| 202 |
-
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 203 |
-
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 204 |
-
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 205 |
-
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 206 |
-
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 207 |
-
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 208 |
-
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 209 |
-
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 210 |
-
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 211 |
-
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 212 |
-
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 213 |
-
</tr>
|
| 214 |
-
|
| 215 |
-
<tr>
|
| 216 |
-
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">Granite-4.0-Hybrid-Small</td>
|
| 217 |
-
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 218 |
-
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 219 |
-
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 220 |
-
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 221 |
-
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 222 |
-
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 223 |
-
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 224 |
-
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 225 |
-
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 226 |
-
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 227 |
-
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 228 |
</tr>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 229 |
</tbody></table>
|
| 230 |
|
|
|
|
| 231 |
<table>
|
| 232 |
-
<caption><b>
|
| 233 |
-
|
| 234 |
-
<tr>
|
| 235 |
-
<th style="text-align:left; background-color: #001d6c; color: white;">Models</th>
|
| 236 |
-
<th style="text-align:center; background-color: #001d6c; color: white;">IFEval</th>
|
| 237 |
-
<th style="text-align:center; background-color: #001d6c; color: white;">BBH</th>
|
| 238 |
-
<th style="text-align:center; background-color: #001d6c; color: white;">MATH Lvl 5</th>
|
| 239 |
-
<th style="text-align:center; background-color: #001d6c; color: white;">GPQA</th>
|
| 240 |
-
<th style="text-align:center; background-color: #001d6c; color: white;">MUSR</th>
|
| 241 |
-
<th style="text-align:center; background-color: #001d6c; color: white;">MMLU-Pro</th>
|
| 242 |
-
<th style="text-align:center; background-color: #001d6c; color: white;">Avg</th>
|
| 243 |
-
</tr></thead>
|
| 244 |
-
<tbody>
|
| 245 |
-
<tr>
|
| 246 |
-
<td style="text-align:left; background-color: #DAE8FF; color: black;">Granite-4.0-H-Tiny</td>
|
| 247 |
-
<td style="text-align:center; background-color: #DAE8FF; color: black;"></td>
|
| 248 |
-
<td style="text-align:center; background-color: #DAE8FF; color: black;"></td>
|
| 249 |
-
<td style="text-align:center; background-color: #DAE8FF; color: black;"></td>
|
| 250 |
-
<td style="text-align:center; background-color: #DAE8FF; color: black;"></td>
|
| 251 |
-
<td style="text-align:center; background-color: #DAE8FF; color: black;"></td>
|
| 252 |
-
<td style="text-align:center; background-color: #DAE8FF; color: black;"></td>
|
| 253 |
-
<td style="text-align:center; background-color: #DAE8FF; color: black;"></td>
|
| 254 |
-
</tr>
|
| 255 |
-
<tr>
|
| 256 |
-
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">Granite-4.0-Hybrid-Micro</td>
|
| 257 |
-
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 258 |
-
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 259 |
-
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 260 |
-
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 261 |
-
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 262 |
-
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 263 |
-
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 264 |
-
</tr>
|
| 265 |
-
<tr>
|
| 266 |
-
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">Granite-4.0-Hybrid-Tiny</td>
|
| 267 |
-
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 268 |
-
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 269 |
-
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 270 |
-
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 271 |
-
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 272 |
-
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 273 |
-
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 274 |
-
</tr>
|
| 275 |
<tr>
|
| 276 |
-
<
|
| 277 |
-
<
|
| 278 |
-
<
|
| 279 |
-
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 280 |
-
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 281 |
-
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 282 |
-
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 283 |
-
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 284 |
</tr>
|
| 285 |
-
</
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 286 |
|
| 287 |
**Model Architecture:**
|
| 288 |
Granite-4.0-H-Tiny baseline is built on a decoder-only MoE transformer architecture. Core components of this architecture are: GQA, Mamba2, MoEs with shared experts, SwiGLU activation, RMSNorm, and shared input/output embeddings.
|
|
|
|
| 33 |
* Code related tasks
|
| 34 |
* Function-calling tasks
|
| 35 |
* Multilingual dialog use cases
|
| 36 |
+
* Fill-In-the-Middle (FIM) code completions
|
| 37 |
|
| 38 |
<!-- <todo>Need to test the examples. (especially the tool calling and RAG ones)</todo>
|
| 39 |
-->
|
|
|
|
| 150 |
*Coming soon* -->
|
| 151 |
|
| 152 |
**Evaluation Results:**
|
|
|
|
| 153 |
|
| 154 |
+
<table>
|
| 155 |
+
<!-- <caption><b> All Results</b></caption> -->
|
| 156 |
<thead>
|
| 157 |
<tr>
|
| 158 |
+
<th style="text-align:left; background-color: #001d6c; color: white;">Benchmarks</th>
|
| 159 |
+
<th style="text-align:left; background-color: #001d6c; color: white;">Metric</th>
|
| 160 |
+
<th style="text-align:center; background-color: #001d6c; color: white;">Micro Dense</th>
|
| 161 |
+
<th style="text-align:center; background-color: #001d6c; color: white;">H Micro Dense</th>
|
| 162 |
+
<th style="text-align:center; background-color: #001d6c; color: white;">H Tiny MoE</th>
|
| 163 |
+
<th style="text-align:center; background-color: #001d6c; color: white;">H Small MoE</th>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 164 |
</tr>
|
| 165 |
+
</thead>
|
| 166 |
+
<tbody>
|
| 167 |
+
<tr>
|
| 168 |
+
<td colspan="6" style="text-align:center; background-color: #FFFFFF; color: #2D2D2D; font-style:italic;">
|
| 169 |
+
General Tasks
|
| 170 |
+
</td>
|
| 171 |
+
</tr>
|
| 172 |
+
<tr>
|
| 173 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">MMLU</td>
|
| 174 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">5-shot</td>
|
| 175 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">65.98</td>
|
| 176 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">67.43</td>
|
| 177 |
+
<td style="text-align:right; background-color: #DAE8FF; color: #2D2D2D;">68.65</td>
|
| 178 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">78.44</td>
|
| 179 |
+
</tr>
|
| 180 |
+
<tr>
|
| 181 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">MMLU-Pro</td>
|
| 182 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">5-shot, CoT</td>
|
| 183 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">44.5</td>
|
| 184 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">43.48</td>
|
| 185 |
+
<td style="text-align:right; background-color: #DAE8FF; color: #2D2D2D;">44.94</td>
|
| 186 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">55.47</td>
|
| 187 |
+
</tr>
|
| 188 |
+
<tr>
|
| 189 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">BBH</td>
|
| 190 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">3-shot, CoT</td>
|
| 191 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">72.48</td>
|
| 192 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">69.36</td>
|
| 193 |
+
<td style="text-align:right; background-color: #DAE8FF; color: #2D2D2D;">66.34</td>
|
| 194 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">81.62</td>
|
| 195 |
+
</tr>
|
| 196 |
+
<tr>
|
| 197 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">AGI EVAL</td>
|
| 198 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">0-shot, CoT</td>
|
| 199 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">64.29</td>
|
| 200 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">59</td>
|
| 201 |
+
<td style="text-align:right; background-color: #DAE8FF; color: #2D2D2D;">62.15</td>
|
| 202 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">70.63</td>
|
| 203 |
+
</tr>
|
| 204 |
+
<tr>
|
| 205 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">GPQA</td>
|
| 206 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">0-shot, CoT</td>
|
| 207 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">30.14</td>
|
| 208 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">32.15</td>
|
| 209 |
+
<td style="text-align:right; background-color: #DAE8FF; color: #2D2D2D;">32.59</td>
|
| 210 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">40.63</td>
|
| 211 |
+
</tr>
|
| 212 |
+
<tr>
|
| 213 |
+
<td colspan="6" style="text-align:center; background-color: #FFFFFF; color: #2D2D2D; font-style:italic;">
|
| 214 |
+
Alignment Tasks
|
| 215 |
+
</td>
|
| 216 |
+
</tr>
|
| 217 |
+
<tr>
|
| 218 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">AlpacaEval 2.0</td>
|
| 219 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 220 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">29.49</td>
|
| 221 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">31.49</td>
|
| 222 |
+
<td style="text-align:right; background-color: #DAE8FF; color: #2D2D2D;">30.61</td>
|
| 223 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">42.48</td>
|
| 224 |
+
</tr>
|
| 225 |
+
<tr>
|
| 226 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">IFEval</td>
|
| 227 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">Instruct, Strict</td>
|
| 228 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">85.5</td>
|
| 229 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">86.94</td>
|
| 230 |
+
<td style="text-align:right; background-color: #DAE8FF; color: #2D2D2D;">84.78</td>
|
| 231 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">89.87</td>
|
| 232 |
+
</tr>
|
| 233 |
+
<tr>
|
| 234 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">IFEval</td>
|
| 235 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">Prompt, Strict</td>
|
| 236 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">79.12</td>
|
| 237 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">81.71</td>
|
| 238 |
+
<td style="text-align:right; background-color: #DAE8FF; color: #2D2D2D;">78.1</td>
|
| 239 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">85.22</td>
|
| 240 |
+
</tr>
|
| 241 |
+
<tr>
|
| 242 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">IFEval</td>
|
| 243 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">Average</td>
|
| 244 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">82.31</td>
|
| 245 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">84.32</td>
|
| 246 |
+
<td style="text-align:right; background-color: #DAE8FF; color: #2D2D2D;">81.44</td>
|
| 247 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">87.55</td>
|
| 248 |
+
</tr>
|
| 249 |
+
<tr>
|
| 250 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">ArenaHard</td>
|
| 251 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 252 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">25.84</td>
|
| 253 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">36.15</td>
|
| 254 |
+
<td style="text-align:right; background-color: #DAE8FF; color: #2D2D2D;">35.75</td>
|
| 255 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">46.48</td>
|
| 256 |
+
</tr>
|
| 257 |
+
<tr>
|
| 258 |
+
<td colspan="6" style="text-align:center; background-color: #FFFFFF; color: #2D2D2D; font-style:italic;">
|
| 259 |
+
Math Tasks
|
| 260 |
+
</td>
|
| 261 |
+
</tr>
|
| 262 |
+
<tr>
|
| 263 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">GSM8K</td>
|
| 264 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">8-shot</td>
|
| 265 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">85.45</td>
|
| 266 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">81.35</td>
|
| 267 |
+
<td style="text-align:right; background-color: #DAE8FF; color: #2D2D2D;">84.69</td>
|
| 268 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">87.27</td>
|
| 269 |
+
</tr>
|
| 270 |
+
<tr>
|
| 271 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">GSM8K Symbolic</td>
|
| 272 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">8-shot</td>
|
| 273 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">79.82</td>
|
| 274 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">77.5</td>
|
| 275 |
+
<td style="text-align:right; background-color: #DAE8FF; color: #2D2D2D;">81.1</td>
|
| 276 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">87.38</td>
|
| 277 |
+
</tr>
|
| 278 |
+
<tr>
|
| 279 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">Minerva Math</td>
|
| 280 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">0-shot, CoT</td>
|
| 281 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">62.06</td>
|
| 282 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">66.44</td>
|
| 283 |
+
<td style="text-align:right; background-color: #DAE8FF; color: #2D2D2D;">69.64</td>
|
| 284 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">74</td>
|
| 285 |
+
</tr>
|
| 286 |
+
<tr>
|
| 287 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">DeepMind Math</td>
|
| 288 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">0-shot, CoT</td>
|
| 289 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">44.56</td>
|
| 290 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">43.83</td>
|
| 291 |
+
<td style="text-align:right; background-color: #DAE8FF; color: #2D2D2D;">49.92</td>
|
| 292 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">59.33</td>
|
| 293 |
+
</tr>
|
| 294 |
+
<tr>
|
| 295 |
+
<td colspan="6" style="text-align:center; background-color: #FFFFFF; color: #2D2D2D; font-style:italic;">
|
| 296 |
+
Code Tasks
|
| 297 |
+
</td>
|
| 298 |
+
</tr>
|
| 299 |
+
<tr>
|
| 300 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">HumanEval</td>
|
| 301 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">pass@1</td>
|
| 302 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">80</td>
|
| 303 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">81</td>
|
| 304 |
+
<td style="text-align:right; background-color: #DAE8FF; color: #2D2D2D;">83</td>
|
| 305 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">88</td>
|
| 306 |
+
</tr>
|
| 307 |
+
<tr>
|
| 308 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">HumanEval+</td>
|
| 309 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">pass@1</td>
|
| 310 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">72</td>
|
| 311 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">75</td>
|
| 312 |
+
<td style="text-align:right; background-color: #DAE8FF; color: #2D2D2D;">76</td>
|
| 313 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">83</td>
|
| 314 |
+
</tr>
|
| 315 |
+
<tr>
|
| 316 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">MBPP</td>
|
| 317 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">pass@1</td>
|
| 318 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">72</td>
|
| 319 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">73</td>
|
| 320 |
+
<td style="text-align:right; background-color: #DAE8FF; color: #2D2D2D;">80</td>
|
| 321 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">84</td>
|
| 322 |
+
</tr>
|
| 323 |
+
<tr>
|
| 324 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">MBPP+</td>
|
| 325 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">pass@1</td>
|
| 326 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">64</td>
|
| 327 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">64</td>
|
| 328 |
+
<td style="text-align:right; background-color: #DAE8FF; color: #2D2D2D;">69</td>
|
| 329 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">71</td>
|
| 330 |
+
</tr>
|
| 331 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">CRUXEval-O</td>
|
| 332 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">pass@1</td>
|
| 333 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">41.5</td>
|
| 334 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">41.25</td>
|
| 335 |
+
<td style="text-align:right; background-color: #DAE8FF; color: #2D2D2D;">39.63</td>
|
| 336 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">50.25</td>
|
| 337 |
+
</tr>
|
| 338 |
+
<tr>
|
| 339 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">BigCodeBench</td>
|
| 340 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">pass@1</td>
|
| 341 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">39.21</td>
|
| 342 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">37.9</td>
|
| 343 |
+
<td style="text-align:right; background-color: #DAE8FF; color: #2D2D2D;">41.06</td>
|
| 344 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">46.23</td>
|
| 345 |
+
</tr>
|
| 346 |
+
<tr>
|
| 347 |
+
<td colspan="6" style="text-align:center; background-color: #FFFFFF; color: #2D2D2D; font-style:italic;">
|
| 348 |
+
Tool Calling Tasks
|
| 349 |
+
</td>
|
| 350 |
+
</tr>
|
| 351 |
+
<tr>
|
| 352 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">BFCL v3</td>
|
| 353 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 354 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">59.98</td>
|
| 355 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">57.56</td>
|
| 356 |
+
<td style="text-align:right; background-color: #DAE8FF; color: #2D2D2D;">57.65</td>
|
| 357 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">64.69</td>
|
| 358 |
+
</tr>
|
| 359 |
+
<tr>
|
| 360 |
+
<td colspan="6" style="text-align:center; background-color: #FFFFFF; color: #2D2D2D; font-style:italic;">
|
| 361 |
+
Multilingual Tasks
|
| 362 |
+
</td>
|
| 363 |
+
</tr>
|
| 364 |
+
<tr>
|
| 365 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">MULTIPLE</td>
|
| 366 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">pass@1</td>
|
| 367 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">49.21</td>
|
| 368 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">49.46</td>
|
| 369 |
+
<td style="text-align:right; background-color: #DAE8FF; color: #2D2D2D;">55.83</td>
|
| 370 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">57.37</td>
|
| 371 |
+
</tr>
|
| 372 |
+
<tr>
|
| 373 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">MMMLU</td>
|
| 374 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">5-shot</td>
|
| 375 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">55.14</td>
|
| 376 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">55.19</td>
|
| 377 |
+
<td style="text-align:right; background-color: #DAE8FF; color: #2D2D2D;">61.87</td>
|
| 378 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">69.69</td>
|
| 379 |
+
</tr>
|
| 380 |
+
<tr>
|
| 381 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">INCLUDE</td>
|
| 382 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">5-shot</td>
|
| 383 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">51.62</td>
|
| 384 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">50.51</td>
|
| 385 |
+
<td style="text-align:right; background-color: #DAE8FF; color: #2D2D2D;">53.12</td>
|
| 386 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">63.97</td>
|
| 387 |
+
</tr>
|
| 388 |
+
<tr>
|
| 389 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">MGSM</td>
|
| 390 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">8-shot</td>
|
| 391 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">28.56</td>
|
| 392 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">44.48</td>
|
| 393 |
+
<td style="text-align:right; background-color: #DAE8FF; color: #2D2D2D;">45.36</td>
|
| 394 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">38.72</td>
|
| 395 |
+
</tr>
|
| 396 |
+
<tr>
|
| 397 |
+
<td colspan="6" style="text-align:center; background-color: #FFFFFF; color: #2D2D2D; font-style:italic;">
|
| 398 |
+
Safety
|
| 399 |
+
</td>
|
| 400 |
+
</tr>
|
| 401 |
+
<tr>
|
| 402 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">SALAD-Bench</td>
|
| 403 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 404 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">97.06</td>
|
| 405 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">96.28</td>
|
| 406 |
+
<td style="text-align:right; background-color: #DAE8FF; color: #2D2D2D;">97.77</td>
|
| 407 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">97.3</td>
|
| 408 |
+
</tr>
|
| 409 |
+
<tr>
|
| 410 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">AttaQ</td>
|
| 411 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;"></td>
|
| 412 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">86.05</td>
|
| 413 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">84.44</td>
|
| 414 |
+
<td style="text-align:right; background-color: #DAE8FF; color: #2D2D2D;">86.61</td>
|
| 415 |
+
<td style="text-align:right; background-color: #FFFFFF; color: #2D2D2D;">86.64</td>
|
| 416 |
+
</tr>
|
| 417 |
</tbody></table>
|
| 418 |
|
| 419 |
+
|
| 420 |
<table>
|
| 421 |
+
<caption><b>Multilingual Benchmarks and thr included languages:</b></caption>
|
| 422 |
+
<thead>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 423 |
<tr>
|
| 424 |
+
<th style="text-align:left; background-color: #001d6c; color: white;">Benchmarks</th>
|
| 425 |
+
<th style="text-align:left; background-color: #001d6c; color: white;"># Langs</th>
|
| 426 |
+
<th style="text-align:center; background-color: #001d6c; color: white;">Languages</th>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 427 |
</tr>
|
| 428 |
+
</thead>
|
| 429 |
+
<tbody>
|
| 430 |
+
<tr>
|
| 431 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">MMMLU</td>
|
| 432 |
+
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;">11</td>
|
| 433 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">ar, de, en, es, fr, ja, ko, pt, zh, bn, hi</td>
|
| 434 |
+
</tr>
|
| 435 |
+
<tr>
|
| 436 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">INCLUDE</td>
|
| 437 |
+
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;">14</td>
|
| 438 |
+
<!-- <td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">hindi, bengali, tamil, telugu, arabic, german, spanish, french, italian, japanese, korean, dutch, portuguese, chinese</td> -->
|
| 439 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">hi, bn, ta, te, ar, de, es, fr, it, ja, ko, nl, pt, zh</td>
|
| 440 |
+
|
| 441 |
+
</tr>
|
| 442 |
+
<tr>
|
| 443 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">MGSM</td>
|
| 444 |
+
<td style="text-align:center; background-color: #FFFFFF; color: #2D2D2D;">5</td>
|
| 445 |
+
<td style="text-align:left; background-color: #FFFFFF; color: #2D2D2D;">en, es, fr, ja, zh</td>
|
| 446 |
+
</tr>
|
| 447 |
+
</tbody>
|
| 448 |
+
</table>
|
| 449 |
|
| 450 |
**Model Architecture:**
|
| 451 |
Granite-4.0-H-Tiny baseline is built on a decoder-only MoE transformer architecture. Core components of this architecture are: GQA, Mamba2, MoEs with shared experts, SwiGLU activation, RMSNorm, and shared input/output embeddings.
|
model.sig
CHANGED
|
@@ -1 +1 @@
|
|
| 1 |
-
{"mediaType": "application/vnd.dev.sigstore.bundle.v0.3+json", "verificationMaterial": {"certificate": {"rawBytes": "
|
|
|
|
| 1 |
+
{"mediaType": "application/vnd.dev.sigstore.bundle.v0.3+json", "verificationMaterial": {"certificate": {"rawBytes": "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"}, "tlogEntries": [{"logIndex": "574894963", "logId": {"keyId": "wNI9atQGlz+VWfO6LRygH4QUfY/8W4RFwiT5i5WRgB0="}, "kindVersion": {"kind": "dsse", "version": "0.0.1"}, "integratedTime": "1759288174", "inclusionPromise": {"signedEntryTimestamp": "MEQCIC8qqOflxiS2uxFV8HBBBZ67dhuKqEg31sJNjH2Ors60AiAJXfjIPwl5laYKutPIvn0VQ0A/cRK+rxA16liTMGU9nQ=="}, "inclusionProof": {"logIndex": "452990701", "rootHash": "49d+YhXmco/GbYcPcMUe7WQVEimu/BwkVoaCnZ7bK8w=", "treeSize": "452990702", "hashes": ["Qlr6QipJ62j8+NBaYTR01Hqw0nIzH1wno4X3ci9kwY0=", "lQ5WUmzEL1ORqd6u5nBFfJ/hdZMjVqeEyPZPnqyt6F4=", "dv24SBlMAzsVB9HNCvp+57Cw1ftefAfKnj/DQ4taOns=", "y5w8mkaJZcPQlPgJOAGL3rtCukQ7exbPQJ9S5XpyjOk=", "gT1Kqt2f8QKuo162Onqr3CgJg6gUulgyiT2iCWbPsNI=", "bVr9zLl3JhPygTZRIQSP7VGxz/JdcHjcHmlFzZBnJAs=", "aIb0AYTWl870/k2CKxxLzLh4Y/pqbXA/ZZ08kcW85PI=", "4ZeUb/c34V9eHBZ8ZVviO0BdbXlSBRhucsphJm5/IM4=", "srFapivtSV0gA1N5LiVWoAVNaXFUlj6zJluZCXK3pa4=", "5J6N5aNEMk6YY8lPNH0zZx95RNICDJ12cMAmpwtvGTo=", "/buh7NyeUdCgJPdaMbTAtiQpMkNHhk+xMsI2shkTieM=", "1mfy94KpcItqshH9+gwqV6jccupcaMpVsF28New8zDY=", "vS7O4ozHIQZJWBiov+mkpI27GE8zAmVCEkRcP3NDyNE="], "checkpoint": {"envelope": "rekor.sigstore.dev - 1193050959916656506\n452990702\n49d+YhXmco/GbYcPcMUe7WQVEimu/BwkVoaCnZ7bK8w=\n\n\u2014 rekor.sigstore.dev wNI9ajBFAiEAtXj6Ak8mFJiOeZoqyF8gcTLxxXCuOAo5Shffk3Yj4YwCIDKtBGnJvA0h5GNPehq/EIcKJWttlDKIR/cSm9oDIyNE\n"}}, "canonicalizedBody": "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"}], "timestampVerificationData": {}}, "dsseEnvelope": {"payload": "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", "payloadType": "application/vnd.in-toto+json", "signatures": [{"sig": "MEMCIG9Xn0Aupy2I7YA7jhkNx8+YU39dxM9PuUkulxj0h+IpAh8BOQkjeayuEp8v0Vv411T4l5Jws1U0nnzQaA7H/s2v"}]}}
|