File size: 7,561 Bytes
ff8fd11 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 | Hardware Resource Needed for RL
===============================
Last updated: 06/25/2025.
Since RL requires more resources compared to regular training,
determining how much resources are needed to successfully run it before training
is a relatively difficult task. To provide more people with reference points for
resource selection when dealing with different models and tasks, this section is
mainly dedicated to introducing the environmental requirements based on experiments
we have conducted.
However, due to limited staff and equipment resources, we also hope for more
contributions from the open-source community. When submitting a PR, it is necessary
to provide a script to be added to the example/tuning scripts.
We need two types of scripts: one is the configuration that can run with the **minimum
resources(min)**, and the other is the configuration that runs with **recommended resources(recommended)**. For the former,
it can be understood as a script that can run after applying all memory optimization techniques
(e.g., offload, gradient checkpointing). For the latter, it can be understood as a script that
can run while avoiding operations that incur additional time overhead as much as possible (targetting best throughput).
When defining script names, please follow this format:
``[model]_[task]_[gpunums]_[device]_[train]_[infer].sh``. This will effectively improve
the script's recognizability. You can place the script under the ``examples/tuning/`` directory.
If you happen to have a configuration that has already been tested, we welcome you to submit
a PR and include a screenshot from Wandb or other verifiable evidence.
----------------------------------------
0.5B
~~~
.. list-table::
:widths: auto
:header-rows: 1
* - Tag
- Model
- Task
- Resource
- MaxBatch
- Train
- Infer
- Link
- Contributor
* - MIN
- Qwen2.5-0.5B
- GRPO-LoRA
- 1*H100
- 116
- fsdp
- vllm0.8.3
- `qwen2-0.5b_grpo-lora_1_h100_fsdp_vllm.sh <https://github.com/volcengine/verl/blob/main/examples/tuning/0.5b/qwen2-0.5b_grpo-lora_1_h100_fsdp_vllm.sh>`_
- `SimonHuang <thelongestusernameofall@gmail.com>`_
1.5B
~~~
.. list-table::
:widths: auto
:header-rows: 1
* - Tag
- Model
- Task
- Resource
- MaxBatch
- Train
- Infer
- Link
- Contributor
* - MIN
- Qwen2.5-1.5B
- GRPO-LoRA
- 1*H100
- 128
- fsdp
- vllm0.8.3
- `qwen2-1.5b_grpo-lora_1_h100_fsdp_vllm.sh <https://github.com/volcengine/verl/blob/main/examples/tuning/1.5b/qwen2-1.5b_grpo-lora_1_h100_fsdp_vllm.sh>`_
- `SimonHuang <thelongestusernameofall@gmail.com>`_
3B
~~~
.. list-table::
:widths: auto
:header-rows: 1
* - Tag
- Model
- Task
- Resource
- MaxBatch
- Train
- Infer
- Link
- Contributor
* - MIN
- Qwen2.5-3B
- GRPO-LoRA
- 1*H100
- 62
- fsdp
- vllm0.8.3
- `qwen2-3b_grpo-lora_1_h100_fsdp_vllm.sh <https://github.com/volcengine/verl/blob/main/examples/tuning/3b/qwen2-3b_grpo-lora_1_h100_fsdp_vllm.sh>`_
- `SimonHuang <thelongestusernameofall@gmail.com>`_
7B
~~~
.. list-table::
:widths: auto
:header-rows: 1
* - Tag
- Model
- Task
- Resource
- MaxBatch
- Train
- Infer
- Link
- Contributor
* - MIN
- Qwen2-7B
- GRPO
- 2*H800
- \
- fsdp
- vllm0.8.2
- `qwen2-7b_grpo_2_h800_fsdp_vllm <https://github.com/volcengine/verl/blob/main/examples/tuning/7b/qwen2-7b_grpo_2_h800_fsdp_vllm.sh>`_
- `Xiangyongan <xiangyongan@bytedance.com>`_
* - MIN
- Qwen2.5-7B
- GRPO-LoRA
- 1*H100
- 16
- fsdp
- vllm0.8.3
- `qwen2-7b_grpo-lora_1_h100_fsdp_vllm.sh <https://github.com/volcengine/verl/blob/main/examples/tuning/7b/qwen2-7b_grpo-lora_1_h100_fsdp_vllm.sh>`_
- `SimonHuang <thelongestusernameofall@gmail.com>`_
14B
~~~
.. list-table::
:widths: auto
:header-rows: 1
* - Tag
- Model
- Task
- Resource
- MaxBatch
- Train
- Infer
- Link
- Contributor
* - MIN
- Qwen2-14B
- GRPO
- 4*H800
- \
- fsdp
- vllm0.8.2
- `qwen2-14b_grpo_4_h800_fsdp_vllm <https://github.com/volcengine/verl/blob/main/examples/tuning/14b/qwen2-14b_grpo_4_h800_fsdp_vllm.sh>`_
- `Xiangyongan <xiangyongan@bytedance.com>`_
* - MIN
- Qwen2.5-14B
- GRPO-LoRA
- 2*H100
- 116
- fsdp
- vllm0.8.3
- `qwen2-14b_grpo-lora_2_h100_fsdp_vllm.sh <https://github.com/volcengine/verl/blob/main/examples/tuning/14b/qwen2-14b_grpo-lora_2_h100_fsdp_vllm.sh>`_
- `SimonHuang <thelongestusernameofall@gmail.com>`_
32B
~~~
.. list-table::
:widths: auto
:header-rows: 1
* - Tag
- Model
- Task
- Resource
- MaxBatch
- Train
- Infer
- Link
- Contributor
* - MIN
- Qwen2-32B
- GRPO
- 8*H20
- \
- megatron
- vllm0.8.2
- `qwen2-32b_grpo_8_h20_megatron_vllm <https://github.com/volcengine/verl/tree/main/examples/tuning/32b/qwen2_32B_grpo_8_h20_megatron_vllm.sh>`_
- `Xiangyongan <xiangyongan@bytedance.com>`_
* - MIN
- Qwen2.5-32B
- GRPO-LoRA
- 4*H100
- 180
- fsdp
- vllm0.8.3
- `qwen2-32b_grpo-lora_4_h100_fsdp_vllm.sh <https://github.com/volcengine/verl/blob/main/examples/tuning/32b/qwen2-32b_grpo-lora_4_h100_fsdp_vllm.sh>`_
- `SimonHuang <thelongestusernameofall@gmail.com>`_
70B
~~~
.. list-table::
:widths: auto
:header-rows: 1
* - Tag
- Model
- Task
- Resource
- MaxBatch
- Train
- Infer
- Link
- Contributor
* - MIN
- Qwen2-70B
- GRPO
- 32*H20
- \
- fsdp
- vllm0.8.2
- `qwen2-70b_grpo_32_h20_fsdp_vllm <https://github.com/volcengine/verl/blob/main/examples/tuning/70b/qwen2-70b_grpo_32_h20_fsdp_vllm.sh>`_
- `Xiangyongan <xiangyongan@bytedance.com>`_
* - MIN
- Qwen2-70B
- GRPO
- 32*H800
- \
- fsdp
- vllm0.8.3
- `qwen2-70b_grpo_32_h800_fsdp_vllm <https://github.com/volcengine/verl/blob/main/examples/tuning/70b/qwen2-70b_grpo_32_h800_fsdp_vllm.sh>`_
- `Xiangyongan <xiangyongan@bytedance.com>`_
* - MIN
- Qwen2.5-72B
- GRPO-LoRA
- 8*H100
- 176
- fsdp
- vllm0.8.3
- `qwen2-72b_grpo-lora_8_h100_fsdp_vllm.sh <https://github.com/volcengine/verl/blob/main/examples/tuning/70b/qwen2-72b_grpo-lora_8_h100_fsdp_vllm.sh>`_
- `SimonHuang <thelongestusernameofall@gmail.com>`_
405B
~~~~
.. table::
:widths: auto
====== ====== ====== ======== ======== ====== ====== ======
tag model task resource MaxBatch train infer link
====== ====== ====== ======== ======== ====== ====== ======
\ \ \ \ \ \ \
====== ====== ====== ======== ======== ====== ====== ======
671B
~~~~
.. table::
:widths: auto
====== ====== ====== ======== ======== ====== ====== ======
tag model task resource MaxBatch train infer link
====== ====== ====== ======== ======== ====== ====== ======
\ \ \ \ \ \ \
====== ====== ====== ======== ======== ====== ====== ======
|