Skip to content

Commit 6240637

Browse files
committed
chore: remove xcomposer support
1 parent 03d4f44 commit 6240637

16 files changed

Lines changed: 16 additions & 725 deletions

File tree

README.md

Lines changed: 3 additions & 5 deletions
Original file line numberDiff line numberDiff line change
@@ -52,11 +52,11 @@ ______________________________________________________________________
5252
- \[2024/09\] LMDeploy PyTorchEngine achieves 1.3x faster on Llama3-8B inference by introducing CUDA graph
5353
- \[2024/08\] LMDeploy is integrated into [modelscope/swift](https://github.com/modelscope/swift) as the default accelerator for VLMs inference
5454
- \[2024/07\] Support Llama3.1 8B, 70B and its TOOLS CALLING
55-
- \[2024/07\] Support [InternVL2](docs/en/multi_modal/internvl.md) full-series models, [InternLM-XComposer2.5](docs/en/multi_modal/xcomposer2d5.md) and [function call](docs/en/llm/api_server_tools.md) of InternLM2.5
55+
- \[2024/07\] Support [InternVL2](docs/en/multi_modal/internvl.md) full-series models and [function call](docs/en/llm/api_server_tools.md) of InternLM2.5
5656
- \[2024/06\] PyTorch engine support DeepSeek-V2 and several VLMs, such as CogVLM2, Mini-InternVL, LlaVA-Next
5757
- \[2024/05\] Balance vision model when deploying VLMs with multiple GPUs
58-
- \[2024/05\] Support 4-bits weight-only quantization and inference on VLMs, such as InternVL v1.5, LLaVa, InternLMXComposer2
59-
- \[2024/04\] Support Llama3 and more VLMs, such as InternVL v1.1, v1.2, MiniGemini, InternLMXComposer2.
58+
- \[2024/05\] Support 4-bits weight-only quantization and inference on VLMs, such as InternVL v1.5 and LLaVa
59+
- \[2024/04\] Support Llama3 and more VLMs, such as InternVL v1.1, v1.2 and MiniGemini.
6060
- \[2024/04\] TurboMind adds online int8/int4 KV cache quantization and inference for all supported devices. Refer [here](docs/en/quantization/kv_quant.md) for detailed guide
6161
- \[2024/04\] TurboMind latest upgrade boosts GQA, rocketing the [internlm2-20b](https://huggingface.co/internlm/internlm2-20b) model inference to 16+ RPS, about 1.8x faster than vLLM.
6262
- \[2024/04\] Support Qwen1.5-MOE and dbrx.
@@ -171,8 +171,6 @@ LMDeploy is a toolkit for compressing, deploying, and serving LLM, developed by
171171
<td>
172172
<ul>
173173
<li>LLaVA(1.5,1.6) (7B-34B)</li>
174-
<li>InternLM-XComposer2 (7B, 4khd-7B)</li>
175-
<li>InternLM-XComposer2.5 (7B)</li>
176174
<li>Qwen-VL (7B)</li>
177175
<li>Qwen2-VL (2B, 7B, 72B)</li>
178176
<li>Qwen2.5-VL (3B, 7B, 72B)</li>

README_ja.md

Lines changed: 3 additions & 5 deletions
Original file line numberDiff line numberDiff line change
@@ -37,11 +37,11 @@ ______________________________________________________________________
3737

3838
- \[2024/08\] 🔥🔥 LMDeployは[modelscope/swift](https://github.com/modelscope/swift)に統合され、VLMs推論のデフォルトアクセラレータとなりました
3939
- \[2024/07\] 🎉🎉 Llama3.1 8B、70Bおよびそのツールコールをサポート
40-
- \[2024/07\] [InternVL2](https://huggingface.co/collections/OpenGVLab/internvl-20-667d3961ab5eb12c7ed1463e)全シリーズモデル、[InternLM-XComposer2.5](docs/en/multi_modal/xcomposer2d5.md)およびInternLM2.5の[ファンクションコール](docs/en/llm/api_server_tools.md)をサポート
40+
- \[2024/07\] [InternVL2](https://huggingface.co/collections/OpenGVLab/internvl-20-667d3961ab5eb12c7ed1463e)全シリーズモデルおよびInternLM2.5の[ファンクションコール](docs/en/llm/api_server_tools.md)をサポート
4141
- \[2024/06\] PyTorchエンジンはDeepSeek-V2およびいくつかのVLMs、例えばCogVLM2、Mini-InternVL、LlaVA-Nextをサポート
4242
- \[2024/05\] 複数のGPUでVLMsをデプロイする際にビジョンモデルをバランスさせる
43-
- \[2024/05\] InternVL v1.5、LLaVa、InternLMXComposer2などのVLMsで4ビットの重みのみの量子化と推論をサポート
44-
- \[2024/04\] Llama3およびInternVL v1.1、v1.2、MiniGemini、InternLMXComposer2などのVLMモデルをサポート
43+
- \[2024/05\] InternVL v1.5、LLaVaなどのVLMsで4ビットの重みのみの量子化と推論をサポート
44+
- \[2024/04\] Llama3およびInternVL v1.1、v1.2、MiniGeminiなどのVLMモデルをサポート
4545
- \[2024/04\] TurboMindはすべてのサポートされているデバイスでのオンラインint8/int4 KVキャッシュ量子化と推論を追加しました。詳細なガイドは[こちら](docs/en/quantization/kv_quant.md)を参照してください
4646
- \[2024/04\] TurboMindの最新アップグレードによりGQAが強化され、[internlm2-20b](https://huggingface.co/internlm/internlm2-20b)モデルの推論が16+ RPSに達し、vLLMの約1.8倍の速さになりました
4747
- \[2024/04\] Qwen1.5-MOEおよびdbrxをサポート
@@ -158,8 +158,6 @@ LMDeploy TurboMindエンジンは卓越した推論能力を持ち、さまざ
158158
<td>
159159
<ul>
160160
<li>LLaVA(1.5,1.6) (7B-34B)</li>
161-
<li>InternLM-XComposer2 (7B, 4khd-7B)</li>
162-
<li>InternLM-XComposer2.5 (7B)</li>
163161
<li>Qwen-VL (7B)</li>
164162
<li>Qwen2-VL (2B, 7B, 72B)</li>
165163
<li>Qwen2.5-VL (3B, 7B, 72B)</li>

README_zh-CN.md

Lines changed: 3 additions & 5 deletions
Original file line numberDiff line numberDiff line change
@@ -52,11 +52,11 @@ ______________________________________________________________________
5252
- \[2024/09\] 通过引入 CUDA Graph,LMDeploy PyTorchEngine 在 Llama3-8B 推理上实现了 1.3 倍的加速
5353
- \[2024/08\] LMDeploy现已集成至 [modelscope/swift](https://github.com/modelscope/swift),成为 VLMs 推理的默认加速引擎
5454
- \[2024/07\] 支持 Llama3.1 8B 和 70B 模型,以及工具调用功能
55-
- \[2024/07\] 支持 [InternVL2](docs/zh_cn/multi_modal/internvl.md) 全系列模型,[InternLM-XComposer2.5](docs/zh_cn/multi_modal/xcomposer2d5.md) 模型和 InternLM2.5 的 [function call 功能](docs/zh_cn/llm/api_server_tools.md)
55+
- \[2024/07\] 支持 [InternVL2](docs/zh_cn/multi_modal/internvl.md) 全系列模型和 InternLM2.5 的 [function call 功能](docs/zh_cn/llm/api_server_tools.md)
5656
- \[2024/06\] PyTorch engine 支持了 DeepSeek-V2 和若干 VLM 模型推理, 比如 CogVLM2,Mini-InternVL,LlaVA-Next
5757
- \[2024/05\] 在多 GPU 上部署 VLM 模型时,支持把视觉部分的模型均分到多卡上
58-
- \[2024/05\] 支持InternVL v1.5, LLaVa, InternLMXComposer2 等 VLMs 模型的 4bit 权重量化和推理
59-
- \[2024/04\] 支持 Llama3 和 InternVL v1.1, v1.2,MiniGemini,InternLM-XComposer2 等 VLM 模型
58+
- \[2024/05\] 支持 InternVL v1.5 和 LLaVa 等 VLMs 模型的 4bit 权重量化和推理
59+
- \[2024/04\] 支持 Llama3 和 InternVL v1.1, v1.2,MiniGemini 等 VLM 模型
6060
- \[2024/04\] TurboMind 支持 kv cache int4/int8 在线量化和推理,适用已支持的所有型号显卡。详情请参考[这里](docs/zh_cn/quantization/kv_quant.md)
6161
- \[2024/04\] TurboMind 引擎升级,优化 GQA 推理。[internlm2-20b](https://huggingface.co/internlm/internlm2-20b) 推理速度达 16+ RPS,约是 vLLM 的 1.8 倍
6262
- \[2024/04\] 支持 Qwen1.5-MOE 和 dbrx.
@@ -173,8 +173,6 @@ LMDeploy TurboMind 引擎拥有卓越的推理能力,在各种规模的模型
173173
<td>
174174
<ul>
175175
<li>LLaVA(1.5,1.6) (7B-34B)</li>
176-
<li>InternLM-XComposer2 (7B, 4khd-7B)</li>
177-
<li>InternLM-XComposer2.5 (7B)</li>
178176
<li>Qwen-VL (7B)</li>
179177
<li>Qwen2-VL (2B, 7B, 72B)</li>
180178
<li>Qwen2.5-VL (3B, 7B, 72B)</li>

autotest/utils/get_run_config.py

Lines changed: 0 additions & 4 deletions
Original file line numberDiff line numberDiff line change
@@ -37,10 +37,6 @@ def get_model_name(model):
3737
return 'internvl-internlm2'
3838
if ('internlm2') in model_name:
3939
return 'internlm2'
40-
if ('internlm-xcomposer2d5') in model_name:
41-
return 'internlm-xcomposer2d5'
42-
if ('internlm-xcomposer2') in model_name:
43-
return 'internlm-xcomposer2'
4440
if ('glm-4') in model_name:
4541
return 'glm4'
4642
if len(model_name.split('-')) > 2 and '-'.join(model_name.split('-')[0:2]) in model_names:

docs/en/multi_modal/index.rst

Lines changed: 0 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -8,7 +8,6 @@ Vision-Language Models
88
deepseek_vl2.md
99
llava.md
1010
internvl.md
11-
xcomposer2d5.md
1211
cogvlm.md
1312
minicpmv.md
1413
phi3.md

docs/en/multi_modal/xcomposer2d5.md

Lines changed: 0 additions & 157 deletions
This file was deleted.

docs/en/supported_models/supported_models.md

Lines changed: 0 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -14,8 +14,6 @@ The following tables detail the models supported by LMDeploy's TurboMind engine
1414
| InternLM2 | 7B - 20B | LLM | Yes | Yes | Yes | Yes |
1515
| InternLM2.5 | 7B | LLM | Yes | Yes | Yes | Yes |
1616
| InternLM3 | 8B | LLM | Yes | Yes | Yes | Yes |
17-
| InternLM-XComposer2 | 7B, 4khd-7B | MLLM | Yes | Yes | Yes | Yes |
18-
| InternLM-XComposer2.5 | 7B | MLLM | Yes | Yes | Yes | Yes |
1917
| Intern-S1 | 241B | MLLM | Yes | Yes | Yes | No |
2018
| Intern-S1-mini | 8.3B | MLLM | Yes | Yes | Yes | No |
2119
| Qwen1.5<sup>\[1\]</sup> | 1.8B - 110B | LLM | Yes | Yes | Yes | Yes |

docs/zh_cn/multi_modal/index.rst

Lines changed: 0 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -8,7 +8,6 @@
88
deepseek_vl2.md
99
llava.md
1010
internvl.md
11-
xcomposer2d5.md
1211
cogvlm.md
1312
minicpmv.md
1413
phi3.md

0 commit comments

Comments
 (0)