Skip to content

FEAT: [Model] Support DeepSeek-V3.1 Quantization and tool #4022

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 3 commits into from
Aug 29, 2025
Merged

FEAT: [Model] Support DeepSeek-V3.1 Quantization and tool #4022

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
13f5b7e
FEAT :Support DeepSeek-V3.1 Quantization and tool
Aug 29, 2025
134147c
FEAT :Support DeepSeek-V3.1 Quantization and tool
Aug 29, 2025
92ed332
Fix:activated_size_in_billions for the DeepSeek V3 series
Aug 29, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
124 changes: 124 additions & 0 deletions xinference/model/llm/llm_family.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4767,6 +4767,7 @@
{
"model_format": "pytorch",
"model_size_in_billions": 671,
"activated_size_in_billions": 37,
"model_src": {
"huggingface": {
"quantizations": [
Expand Down Expand Up @@ -4846,6 +4847,7 @@
{
"model_format": "pytorch",
"model_size_in_billions": 671,
"activated_size_in_billions": 37,
"model_src": {
"huggingface": {
"quantizations": [
Expand All @@ -4866,6 +4868,7 @@
{
"model_format": "awq",
"model_size_in_billions": 671,
"activated_size_in_billions": 37,
"model_src": {
"huggingface": {
"quantizations": [
Expand All @@ -4885,6 +4888,7 @@
{
"model_format": "ggufv2",
"model_size_in_billions": 671,
"activated_size_in_billions": 37,
"model_src": {
"huggingface": {
"quantizations": [
Expand Down Expand Up @@ -5215,6 +5219,7 @@
{
"model_format": "mlx",
"model_size_in_billions": 671,
"activated_size_in_billions": 37,
"model_src": {
"huggingface": {
"quantizations": [
Expand Down Expand Up @@ -5263,6 +5268,7 @@
{
"model_format": "pytorch",
"model_size_in_billions": 671,
"activated_size_in_billions": 37,
"model_src": {
"huggingface": {
"quantizations": [
Expand All @@ -5281,6 +5287,7 @@
{
"model_format": "gptq",
"model_size_in_billions": 671,
"activated_size_in_billions": 37,
"model_src": {
"huggingface": {
"quantizations": [
Expand Down Expand Up @@ -5311,6 +5318,116 @@
"reasoning_start_tag": "<think>",
"reasoning_end_tag": "</think>"
},
{
"version": 2,
"context_length": 131072,
"model_name": "Deepseek-V3.1",
"model_lang": [
"en",
"zh"
],
"model_ability": [
"chat",
"reasoning",
"hybrid",
"tools"
],
"model_description": "DeepSeek-V3.1 is a hybrid model that supports both thinking mode and non-thinking mode.",
"model_specs": [
{
"model_format": "pytorch",
"model_size_in_billions": 671,
"activated_size_in_billions": 37,
"model_src": {
"huggingface": {
"quantizations": [
"none"
],
"model_id": "deepseek-ai/DeepSeek-V3.1"
},
"modelscope": {
"quantizations": [
"none"
],
"model_id": "deepseek-ai/DeepSeek-V3.1"
}
}
},
{
"model_format": "gptq",
"model_size_in_billions": 671,
"activated_size_in_billions": 37,
"model_src": {
"huggingface": {
"quantizations": [
"Int4"
],
"model_id": "cpatonn/DeepSeek-V3.1-GPTQ-4bit"
},
"modelscope": {
"quantizations": [
"Int4"
],
"model_id": "cpatonn/DeepSeek-V3.1-GPTQ-4bit"
}
}
},
{
"model_format": "awq",
"model_size_in_billions": 671,
"activated_size_in_billions": 37,
"model_src": {
"huggingface": {
"quantizations": [
"Int4"
],
"model_id": "QuantTrio/DeepSeek-V3.1-AWQ"
},
"modelscope": {
"quantizations": [
"Int4"
],
"model_id": "tclf90/DeepSeek-V3.1-AWQ"
}
}
},
{
"model_format": "mlx",
"model_size_in_billions": 671,
"activated_size_in_billions": 37,
"model_src": {
"huggingface": {
"quantizations": [
"8bit",
"4bit"
],
"model_id": "mlx-community/DeepSeek-V3.1-{quantization}"
},
"modelscope": {
"quantizations": [
"8bit",
"4bit"
],
"model_id": "mlx-community/DeepSeek-V3.1-{quantization}"
}
}
}
],
"chat_template": "{% if not add_generation_prompt is defined %}{% set add_generation_prompt = false %}{% endif %}{% if not thinking is defined %}{% set thinking = false %}{% endif %}{% set ns = namespace(is_first=false, is_tool=false, system_prompt='', is_first_sp=true, is_last_user=false) %}{%- for message in messages %}{%- if message['role'] == 'system' %}{%- if ns.is_first_sp %}{% set ns.system_prompt = ns.system_prompt + message['content'] %}{% set ns.is_first_sp = false %}{%- else %}{% set ns.system_prompt = ns.system_prompt + '\n\n' + message['content'] %}{%- endif %}{%- endif %}{%- endfor %}{{ bos_token }}{{ ns.system_prompt }}{%- for message in messages %}{%- if message['role'] == 'user' %}{%- set ns.is_tool = false -%}{%- set ns.is_first = false -%}{%- set ns.is_last_user = true -%}{{'<｜User｜>' + message['content']}}{%- endif %}{%- if message['role'] == 'assistant' and message['tool_calls'] is defined and message['tool_calls'] is not none %}{%- if ns.is_last_user %}{{'<｜Assistant｜></think>'}}{%- endif %}{%- set ns.is_last_user = false -%}{%- set ns.is_first = false %}{%- set ns.is_tool = false -%}{%- for tool in message['tool_calls'] %}{%- if not ns.is_first %}{%- if message['content'] is none %}{{'<｜tool▁calls▁begin｜><｜tool▁call▁begin｜>'+ tool['function']['name'] + '<｜tool▁sep｜>' + tool['function']['arguments'] + '<｜tool▁call▁end｜>'}}{%- else %}{{message['content'] + '<｜tool▁calls▁begin｜><｜tool▁call▁begin｜>' + tool['function']['name'] + '<｜tool▁sep｜>' + tool['function']['arguments'] + '<｜tool▁call▁end｜>'}}{%- endif %}{%- set ns.is_first = true -%}{%- else %}{{'<｜tool▁call▁begin｜>'+ tool['function']['name'] + '<｜tool▁sep｜>' + tool['function']['arguments'] + '<｜tool▁call▁end｜>'}}{%- endif %}{%- endfor %}{{'<｜tool▁calls▁end｜><｜end▁of▁sentence｜>'}}{%- endif %}{%- if message['role'] == 'assistant' and (message['tool_calls'] is not defined or message['tool_calls'] is none) %}{%- if ns.is_last_user %}{{'<｜Assistant｜>'}}{%- if message['prefix'] is defined and message['prefix'] and thinking %}{{'<think>'}} {%- else %}{{'</think>'}}{%- endif %}{%- endif %}{%- set ns.is_last_user = false -%}{%- if ns.is_tool %}{{message['content'] + '<｜end▁of▁sentence｜>'}}{%- set ns.is_tool = false -%}{%- else %}{%- set content = message['content'] -%}{%- if '</think>' in content %}{%- set content = content.split('</think>', 1)[1] -%}{%- endif %}{{content + '<｜end▁of▁sentence｜>'}}{%- endif %}{%- endif %}{%- if message['role'] == 'tool' %}{%- set ns.is_last_user = false -%}{%- set ns.is_tool = true -%}{{'<｜tool▁output▁begin｜>' + message['content'] + '<｜tool▁output▁end｜>'}}{%- endif %}{%- endfor -%}{%- if add_generation_prompt and ns.is_last_user and not ns.is_tool %}{{'<｜Assistant｜>'}}{%- if not thinking %}{{'</think>'}}{%- else %}{{'<think>'}}{%- endif %}{% endif %}",
"stop_token_ids": [
1
],
"stop": [
"<｜end▁of▁sentence｜>"
],
"reasoning_start_tag": "<think>",
"reasoning_end_tag": "</think>",
"virtualenv": {
"packages": [
"transformers==4.53.0"
]
}
},
{
"version": 2,
"context_length": 131072,
Expand Down Expand Up @@ -6242,6 +6359,7 @@
{
"model_format": "pytorch",
"model_size_in_billions": 671,
"activated_size_in_billions": 37,
"model_src": {
"huggingface": {
"quantizations": [
Expand All @@ -6262,6 +6380,7 @@
{
"model_format": "awq",
"model_size_in_billions": 671,
"activated_size_in_billions": 37,
"model_src": {
"huggingface": {
"quantizations": [
Expand All @@ -6281,6 +6400,7 @@
{
"model_format": "ggufv2",
"model_size_in_billions": 671,
"activated_size_in_billions": 37,
"model_src": {
"huggingface": {
"quantizations": [
Expand Down Expand Up @@ -6475,6 +6595,7 @@
{
"model_format": "mlx",
"model_size_in_billions": 671,
"activated_size_in_billions": 37,
"model_src": {
"huggingface": {
"quantizations": [
Expand Down Expand Up @@ -6517,6 +6638,7 @@
{
"model_format": "pytorch",
"model_size_in_billions": 671,
"activated_size_in_billions": 37,
"model_src": {
"huggingface": {
"quantizations": [
Expand All @@ -6535,6 +6657,7 @@
{
"model_format": "awq",
"model_size_in_billions": 671,
"activated_size_in_billions": 37,
"model_src": {
"huggingface": {
"quantizations": [
Expand All @@ -6553,6 +6676,7 @@
{
"model_format": "mlx",
"model_size_in_billions": 671,
"activated_size_in_billions": 37,
"model_src": {
"huggingface": {
"quantizations": [
Expand Down
2 changes: 1 addition & 1 deletion xinference/model/llm/utils.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -82,7 +82,7 @@
"HuatuoGPT-o1-LLaMA-3.1",
]

DEEPSEEK_TOOL_CALL_FAMILY = ["deepseek-v3", "deepseek-r1-0528"]
DEEPSEEK_TOOL_CALL_FAMILY = ["deepseek-v3", "deepseek-r1-0528", "Deepseek-V3.1"]

TOOL_CALL_FAMILY = (
QWEN_TOOL_CALL_FAMILY
Expand Down
1 change: 1 addition & 0 deletions xinference/model/llm/vllm/core.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -273,6 +273,7 @@ class VLLMGenerateConfig(TypedDict, total=False):
VLLM_SUPPORTED_CHAT_MODELS.append("Qwen3-Instruct")
VLLM_SUPPORTED_CHAT_MODELS.append("Qwen3-Thinking")
VLLM_SUPPORTED_CHAT_MODELS.append("Qwen3-Coder")
VLLM_SUPPORTED_CHAT_MODELS.append("Deepseek-V3.1")

if VLLM_INSTALLED and VLLM_VERSION >= version.parse("0.10.0"):
VLLM_SUPPORTED_CHAT_MODELS.append("glm-4.5")
Expand Down
Loading