diff --git a/api/core/model_runtime/model_providers/openrouter/llm/deepseek-chat.yaml b/api/core/model_runtime/model_providers/openrouter/llm/deepseek-chat.yaml index 7a1dea695..6743bfcad 100644 --- a/api/core/model_runtime/model_providers/openrouter/llm/deepseek-chat.yaml +++ b/api/core/model_runtime/model_providers/openrouter/llm/deepseek-chat.yaml @@ -35,6 +35,15 @@ parameter_rules: help: zh_Hans: 控制生成结果的随机性。数值越小，随机性越弱；数值越大，随机性越强。一般而言，top_p 和 temperature 两个参数选择一个进行调整即可。 en_US: Control the randomness of generated results. The smaller the value, the weaker the randomness; the larger the value, the stronger the randomness. Generally speaking, you can adjust one of the two parameters top_p and temperature. + - name: top_k + label: + zh_Hans: 取样数量 + en_US: Top k + type: int + help: + zh_Hans: 仅从每个后续标记的前 K 个选项中采样。 + en_US: Only sample from the top K options for each subsequent token. + required: false - name: frequency_penalty use_template: frequency_penalty default: 0 diff --git a/api/core/model_runtime/model_providers/openrouter/llm/deepseek-coder.yaml b/api/core/model_runtime/model_providers/openrouter/llm/deepseek-coder.yaml index c05f4769b..375a4d2d5 100644 --- a/api/core/model_runtime/model_providers/openrouter/llm/deepseek-coder.yaml +++ b/api/core/model_runtime/model_providers/openrouter/llm/deepseek-coder.yaml @@ -18,6 +18,15 @@ parameter_rules: min: 0 max: 1 default: 1 + - name: top_k + label: + zh_Hans: 取样数量 + en_US: Top k + type: int + help: + zh_Hans: 仅从每个后续标记的前 K 个选项中采样。 + en_US: Only sample from the top K options for each subsequent token. + required: false - name: max_tokens use_template: max_tokens min: 1 diff --git a/api/core/model_runtime/model_providers/openrouter/llm/gpt-3.5-turbo.yaml b/api/core/model_runtime/model_providers/openrouter/llm/gpt-3.5-turbo.yaml index 186c1cc66..621ecf065 100644 --- a/api/core/model_runtime/model_providers/openrouter/llm/gpt-3.5-turbo.yaml +++ b/api/core/model_runtime/model_providers/openrouter/llm/gpt-3.5-turbo.yaml @@ -14,6 +14,15 @@ parameter_rules: use_template: temperature - name: top_p use_template: top_p + - name: top_k + label: + zh_Hans: 取样数量 + en_US: Top k + type: int + help: + zh_Hans: 仅从每个后续标记的前 K 个选项中采样。 + en_US: Only sample from the top K options for each subsequent token. + required: false - name: presence_penalty use_template: presence_penalty - name: frequency_penalty diff --git a/api/core/model_runtime/model_providers/openrouter/llm/gpt-4-32k.yaml b/api/core/model_runtime/model_providers/openrouter/llm/gpt-4-32k.yaml index 8c2989b30..887e6d60f 100644 --- a/api/core/model_runtime/model_providers/openrouter/llm/gpt-4-32k.yaml +++ b/api/core/model_runtime/model_providers/openrouter/llm/gpt-4-32k.yaml @@ -14,6 +14,15 @@ parameter_rules: use_template: temperature - name: top_p use_template: top_p + - name: top_k + label: + zh_Hans: 取样数量 + en_US: Top k + type: int + help: + zh_Hans: 仅从每个后续标记的前 K 个选项中采样。 + en_US: Only sample from the top K options for each subsequent token. + required: false - name: presence_penalty use_template: presence_penalty - name: frequency_penalty diff --git a/api/core/model_runtime/model_providers/openrouter/llm/gpt-4.yaml b/api/core/model_runtime/model_providers/openrouter/llm/gpt-4.yaml index ef19d4f6f..66d1f9ae6 100644 --- a/api/core/model_runtime/model_providers/openrouter/llm/gpt-4.yaml +++ b/api/core/model_runtime/model_providers/openrouter/llm/gpt-4.yaml @@ -14,6 +14,15 @@ parameter_rules: use_template: temperature - name: top_p use_template: top_p + - name: top_k + label: + zh_Hans: 取样数量 + en_US: Top k + type: int + help: + zh_Hans: 仅从每个后续标记的前 K 个选项中采样。 + en_US: Only sample from the top K options for each subsequent token. + required: false - name: presence_penalty use_template: presence_penalty - name: frequency_penalty diff --git a/api/core/model_runtime/model_providers/openrouter/llm/gpt-4o-2024-08-06.yaml b/api/core/model_runtime/model_providers/openrouter/llm/gpt-4o-2024-08-06.yaml index 0be325f55..695cc3eed 100644 --- a/api/core/model_runtime/model_providers/openrouter/llm/gpt-4o-2024-08-06.yaml +++ b/api/core/model_runtime/model_providers/openrouter/llm/gpt-4o-2024-08-06.yaml @@ -16,6 +16,15 @@ parameter_rules: use_template: temperature - name: top_p use_template: top_p + - name: top_k + label: + zh_Hans: 取样数量 + en_US: Top k + type: int + help: + zh_Hans: 仅从每个后续标记的前 K 个选项中采样。 + en_US: Only sample from the top K options for each subsequent token. + required: false - name: presence_penalty use_template: presence_penalty - name: frequency_penalty diff --git a/api/core/model_runtime/model_providers/openrouter/llm/gpt-4o-mini.yaml b/api/core/model_runtime/model_providers/openrouter/llm/gpt-4o-mini.yaml index 3b1d95643..e1e588908 100644 --- a/api/core/model_runtime/model_providers/openrouter/llm/gpt-4o-mini.yaml +++ b/api/core/model_runtime/model_providers/openrouter/llm/gpt-4o-mini.yaml @@ -15,6 +15,15 @@ parameter_rules: use_template: temperature - name: top_p use_template: top_p + - name: top_k + label: + zh_Hans: 取样数量 + en_US: Top k + type: int + help: + zh_Hans: 仅从每个后续标记的前 K 个选项中采样。 + en_US: Only sample from the top K options for each subsequent token. + required: false - name: presence_penalty use_template: presence_penalty - name: frequency_penalty diff --git a/api/core/model_runtime/model_providers/openrouter/llm/gpt-4o.yaml b/api/core/model_runtime/model_providers/openrouter/llm/gpt-4o.yaml index a8c97efdd..560bf9d7d 100644 --- a/api/core/model_runtime/model_providers/openrouter/llm/gpt-4o.yaml +++ b/api/core/model_runtime/model_providers/openrouter/llm/gpt-4o.yaml @@ -15,6 +15,15 @@ parameter_rules: use_template: temperature - name: top_p use_template: top_p + - name: top_k + label: + zh_Hans: 取样数量 + en_US: Top k + type: int + help: + zh_Hans: 仅从每个后续标记的前 K 个选项中采样。 + en_US: Only sample from the top K options for each subsequent token. + required: false - name: presence_penalty use_template: presence_penalty - name: frequency_penalty diff --git a/api/core/model_runtime/model_providers/openrouter/llm/llama-3-70b-instruct.yaml b/api/core/model_runtime/model_providers/openrouter/llm/llama-3-70b-instruct.yaml index b91c39e72..04a4a90c6 100644 --- a/api/core/model_runtime/model_providers/openrouter/llm/llama-3-70b-instruct.yaml +++ b/api/core/model_runtime/model_providers/openrouter/llm/llama-3-70b-instruct.yaml @@ -10,6 +10,15 @@ parameter_rules: use_template: temperature - name: top_p use_template: top_p + - name: top_k + label: + zh_Hans: 取样数量 + en_US: Top k + type: int + help: + zh_Hans: 仅从每个后续标记的前 K 个选项中采样。 + en_US: Only sample from the top K options for each subsequent token. + required: false - name: max_tokens use_template: max_tokens required: true diff --git a/api/core/model_runtime/model_providers/openrouter/llm/llama-3-8b-instruct.yaml b/api/core/model_runtime/model_providers/openrouter/llm/llama-3-8b-instruct.yaml index 84b2c7fac..066949d43 100644 --- a/api/core/model_runtime/model_providers/openrouter/llm/llama-3-8b-instruct.yaml +++ b/api/core/model_runtime/model_providers/openrouter/llm/llama-3-8b-instruct.yaml @@ -10,6 +10,15 @@ parameter_rules: use_template: temperature - name: top_p use_template: top_p + - name: top_k + label: + zh_Hans: 取样数量 + en_US: Top k + type: int + help: + zh_Hans: 仅从每个后续标记的前 K 个选项中采样。 + en_US: Only sample from the top K options for each subsequent token. + required: false - name: max_tokens use_template: max_tokens required: true diff --git a/api/core/model_runtime/model_providers/openrouter/llm/llama-3.1-405b-instruct.yaml b/api/core/model_runtime/model_providers/openrouter/llm/llama-3.1-405b-instruct.yaml index a489ce1b5..0cd89dea7 100644 --- a/api/core/model_runtime/model_providers/openrouter/llm/llama-3.1-405b-instruct.yaml +++ b/api/core/model_runtime/model_providers/openrouter/llm/llama-3.1-405b-instruct.yaml @@ -10,6 +10,15 @@ parameter_rules: use_template: temperature - name: top_p use_template: top_p + - name: top_k + label: + zh_Hans: 取样数量 + en_US: Top k + type: int + help: + zh_Hans: 仅从每个后续标记的前 K 个选项中采样。 + en_US: Only sample from the top K options for each subsequent token. + required: false - name: max_tokens use_template: max_tokens required: true diff --git a/api/core/model_runtime/model_providers/openrouter/llm/llama-3.1-70b-instruct.yaml b/api/core/model_runtime/model_providers/openrouter/llm/llama-3.1-70b-instruct.yaml index 12037411b..768ab5ecb 100644 --- a/api/core/model_runtime/model_providers/openrouter/llm/llama-3.1-70b-instruct.yaml +++ b/api/core/model_runtime/model_providers/openrouter/llm/llama-3.1-70b-instruct.yaml @@ -10,6 +10,15 @@ parameter_rules: use_template: temperature - name: top_p use_template: top_p + - name: top_k + label: + zh_Hans: 取样数量 + en_US: Top k + type: int + help: + zh_Hans: 仅从每个后续标记的前 K 个选项中采样。 + en_US: Only sample from the top K options for each subsequent token. + required: false - name: max_tokens use_template: max_tokens required: true diff --git a/api/core/model_runtime/model_providers/openrouter/llm/llama-3.1-8b-instruct.yaml b/api/core/model_runtime/model_providers/openrouter/llm/llama-3.1-8b-instruct.yaml index 6f06493f2..67b6b82b5 100644 --- a/api/core/model_runtime/model_providers/openrouter/llm/llama-3.1-8b-instruct.yaml +++ b/api/core/model_runtime/model_providers/openrouter/llm/llama-3.1-8b-instruct.yaml @@ -10,6 +10,15 @@ parameter_rules: use_template: temperature - name: top_p use_template: top_p + - name: top_k + label: + zh_Hans: 取样数量 + en_US: Top k + type: int + help: + zh_Hans: 仅从每个后续标记的前 K 个选项中采样。 + en_US: Only sample from the top K options for each subsequent token. + required: false - name: max_tokens use_template: max_tokens required: true diff --git a/api/core/model_runtime/model_providers/openrouter/llm/mistral-7b-instruct.yaml b/api/core/model_runtime/model_providers/openrouter/llm/mistral-7b-instruct.yaml index 012dfc55c..d08c016e9 100644 --- a/api/core/model_runtime/model_providers/openrouter/llm/mistral-7b-instruct.yaml +++ b/api/core/model_runtime/model_providers/openrouter/llm/mistral-7b-instruct.yaml @@ -18,6 +18,15 @@ parameter_rules: default: 1 min: 0 max: 1 + - name: top_k + label: + zh_Hans: 取样数量 + en_US: Top k + type: int + help: + zh_Hans: 仅从每个后续标记的前 K 个选项中采样。 + en_US: Only sample from the top K options for each subsequent token. + required: false - name: max_tokens use_template: max_tokens default: 1024 diff --git a/api/core/model_runtime/model_providers/openrouter/llm/mixtral-8x22b-instruct.yaml b/api/core/model_runtime/model_providers/openrouter/llm/mixtral-8x22b-instruct.yaml index f4eb4e45d..e3af0e64d 100644 --- a/api/core/model_runtime/model_providers/openrouter/llm/mixtral-8x22b-instruct.yaml +++ b/api/core/model_runtime/model_providers/openrouter/llm/mixtral-8x22b-instruct.yaml @@ -18,6 +18,15 @@ parameter_rules: default: 1 min: 0 max: 1 + - name: top_k + label: + zh_Hans: 取样数量 + en_US: Top k + type: int + help: + zh_Hans: 仅从每个后续标记的前 K 个选项中采样。 + en_US: Only sample from the top K options for each subsequent token. + required: false - name: max_tokens use_template: max_tokens default: 1024 diff --git a/api/core/model_runtime/model_providers/openrouter/llm/mixtral-8x7b-instruct.yaml b/api/core/model_runtime/model_providers/openrouter/llm/mixtral-8x7b-instruct.yaml index 7871e1f7a..095ea5a85 100644 --- a/api/core/model_runtime/model_providers/openrouter/llm/mixtral-8x7b-instruct.yaml +++ b/api/core/model_runtime/model_providers/openrouter/llm/mixtral-8x7b-instruct.yaml @@ -19,6 +19,15 @@ parameter_rules: default: 1 min: 0 max: 1 + - name: top_k + label: + zh_Hans: 取样数量 + en_US: Top k + type: int + help: + zh_Hans: 仅从每个后续标记的前 K 个选项中采样。 + en_US: Only sample from the top K options for each subsequent token. + required: false - name: max_tokens use_template: max_tokens default: 1024 diff --git a/api/core/model_runtime/model_providers/openrouter/llm/o1-mini.yaml b/api/core/model_runtime/model_providers/openrouter/llm/o1-mini.yaml index 85a918ff5..f4202ee81 100644 --- a/api/core/model_runtime/model_providers/openrouter/llm/o1-mini.yaml +++ b/api/core/model_runtime/model_providers/openrouter/llm/o1-mini.yaml @@ -12,6 +12,15 @@ parameter_rules: use_template: temperature - name: top_p use_template: top_p + - name: top_k + label: + zh_Hans: 取样数量 + en_US: Top k + type: int + help: + zh_Hans: 仅从每个后续标记的前 K 个选项中采样。 + en_US: Only sample from the top K options for each subsequent token. + required: false - name: presence_penalty use_template: presence_penalty - name: frequency_penalty diff --git a/api/core/model_runtime/model_providers/openrouter/llm/o1-preview.yaml b/api/core/model_runtime/model_providers/openrouter/llm/o1-preview.yaml index 74b0a511b..1281b8428 100644 --- a/api/core/model_runtime/model_providers/openrouter/llm/o1-preview.yaml +++ b/api/core/model_runtime/model_providers/openrouter/llm/o1-preview.yaml @@ -12,6 +12,15 @@ parameter_rules: use_template: temperature - name: top_p use_template: top_p + - name: top_k + label: + zh_Hans: 取样数量 + en_US: Top k + type: int + help: + zh_Hans: 仅从每个后续标记的前 K 个选项中采样。 + en_US: Only sample from the top K options for each subsequent token. + required: false - name: presence_penalty use_template: presence_penalty - name: frequency_penalty diff --git a/api/core/model_runtime/model_providers/openrouter/llm/qwen2-72b-instruct.yaml b/api/core/model_runtime/model_providers/openrouter/llm/qwen2-72b-instruct.yaml index 7b75fcb0c..b6058138d 100644 --- a/api/core/model_runtime/model_providers/openrouter/llm/qwen2-72b-instruct.yaml +++ b/api/core/model_runtime/model_providers/openrouter/llm/qwen2-72b-instruct.yaml @@ -21,6 +21,15 @@ parameter_rules: en_US: Specifies the upper limit on the length of generated results. If the generated results are truncated, you can increase this parameter. - name: top_p use_template: top_p + - name: top_k + label: + zh_Hans: 取样数量 + en_US: Top k + type: int + help: + zh_Hans: 仅从每个后续标记的前 K 个选项中采样。 + en_US: Only sample from the top K options for each subsequent token. + required: false - name: frequency_penalty use_template: frequency_penalty pricing: diff --git a/api/core/model_runtime/model_providers/openrouter/llm/qwen2.5-72b-instruct.yaml b/api/core/model_runtime/model_providers/openrouter/llm/qwen2.5-72b-instruct.yaml index f141a40a0..5392b1116 100644 --- a/api/core/model_runtime/model_providers/openrouter/llm/qwen2.5-72b-instruct.yaml +++ b/api/core/model_runtime/model_providers/openrouter/llm/qwen2.5-72b-instruct.yaml @@ -21,6 +21,15 @@ parameter_rules: en_US: Specifies the upper limit on the length of generated results. If the generated results are truncated, you can increase this parameter. - name: top_p use_template: top_p + - name: top_k + label: + zh_Hans: 取样数量 + en_US: Top k + type: int + help: + zh_Hans: 仅从每个后续标记的前 K 个选项中采样。 + en_US: Only sample from the top K options for each subsequent token. + required: false - name: frequency_penalty use_template: frequency_penalty pricing: diff --git a/api/core/model_runtime/model_providers/siliconflow/llm/deepdeek-coder-v2-instruct.yaml b/api/core/model_runtime/model_providers/siliconflow/llm/deepdeek-coder-v2-instruct.yaml index d4431179e..d5f23776e 100644 --- a/api/core/model_runtime/model_providers/siliconflow/llm/deepdeek-coder-v2-instruct.yaml +++ b/api/core/model_runtime/model_providers/siliconflow/llm/deepdeek-coder-v2-instruct.yaml @@ -21,6 +21,15 @@ parameter_rules: en_US: Specifies the upper limit on the length of generated results. If the generated results are truncated, you can increase this parameter. - name: top_p use_template: top_p + - name: top_k + label: + zh_Hans: 取样数量 + en_US: Top k + type: int + help: + zh_Hans: 仅从每个后续标记的前 K 个选项中采样。 + en_US: Only sample from the top K options for each subsequent token. + required: false - name: frequency_penalty use_template: frequency_penalty pricing: diff --git a/api/core/model_runtime/model_providers/siliconflow/llm/deepseek-v2-chat.yaml b/api/core/model_runtime/model_providers/siliconflow/llm/deepseek-v2-chat.yaml index caa6508b5..7aa684ef3 100644 --- a/api/core/model_runtime/model_providers/siliconflow/llm/deepseek-v2-chat.yaml +++ b/api/core/model_runtime/model_providers/siliconflow/llm/deepseek-v2-chat.yaml @@ -21,6 +21,15 @@ parameter_rules: en_US: Specifies the upper limit on the length of generated results. If the generated results are truncated, you can increase this parameter. - name: top_p use_template: top_p + - name: top_k + label: + zh_Hans: 取样数量 + en_US: Top k + type: int + help: + zh_Hans: 仅从每个后续标记的前 K 个选项中采样。 + en_US: Only sample from the top K options for each subsequent token. + required: false - name: frequency_penalty use_template: frequency_penalty pricing: diff --git a/api/core/model_runtime/model_providers/siliconflow/llm/deepseek-v2.5.yaml b/api/core/model_runtime/model_providers/siliconflow/llm/deepseek-v2.5.yaml index 1c8e15ae5..b30fa3e2d 100644 --- a/api/core/model_runtime/model_providers/siliconflow/llm/deepseek-v2.5.yaml +++ b/api/core/model_runtime/model_providers/siliconflow/llm/deepseek-v2.5.yaml @@ -21,6 +21,15 @@ parameter_rules: en_US: Specifies the upper limit on the length of generated results. If the generated results are truncated, you can increase this parameter. - name: top_p use_template: top_p + - name: top_k + label: + zh_Hans: 取样数量 + en_US: Top k + type: int + help: + zh_Hans: 仅从每个后续标记的前 K 个选项中采样。 + en_US: Only sample from the top K options for each subsequent token. + required: false - name: frequency_penalty use_template: frequency_penalty pricing: diff --git a/api/core/model_runtime/model_providers/siliconflow/llm/gemma-2-27b-it.yaml b/api/core/model_runtime/model_providers/siliconflow/llm/gemma-2-27b-it.yaml index 2840e3dcf..f2a1f64bf 100644 --- a/api/core/model_runtime/model_providers/siliconflow/llm/gemma-2-27b-it.yaml +++ b/api/core/model_runtime/model_providers/siliconflow/llm/gemma-2-27b-it.yaml @@ -21,6 +21,15 @@ parameter_rules: en_US: Specifies the upper limit on the length of generated results. If the generated results are truncated, you can increase this parameter. - name: top_p use_template: top_p + - name: top_k + label: + zh_Hans: 取样数量 + en_US: Top k + type: int + help: + zh_Hans: 仅从每个后续标记的前 K 个选项中采样。 + en_US: Only sample from the top K options for each subsequent token. + required: false - name: frequency_penalty use_template: frequency_penalty pricing: diff --git a/api/core/model_runtime/model_providers/siliconflow/llm/gemma-2-9b-it.yaml b/api/core/model_runtime/model_providers/siliconflow/llm/gemma-2-9b-it.yaml index d7e19b46f..b096b9b64 100644 --- a/api/core/model_runtime/model_providers/siliconflow/llm/gemma-2-9b-it.yaml +++ b/api/core/model_runtime/model_providers/siliconflow/llm/gemma-2-9b-it.yaml @@ -21,6 +21,15 @@ parameter_rules: en_US: Specifies the upper limit on the length of generated results. If the generated results are truncated, you can increase this parameter. - name: top_p use_template: top_p + - name: top_k + label: + zh_Hans: 取样数量 + en_US: Top k + type: int + help: + zh_Hans: 仅从每个后续标记的前 K 个选项中采样。 + en_US: Only sample from the top K options for each subsequent token. + required: false - name: frequency_penalty use_template: frequency_penalty pricing: diff --git a/api/core/model_runtime/model_providers/siliconflow/llm/glm4-9b-chat.yaml b/api/core/model_runtime/model_providers/siliconflow/llm/glm4-9b-chat.yaml index 9b32a0247..87acc557b 100644 --- a/api/core/model_runtime/model_providers/siliconflow/llm/glm4-9b-chat.yaml +++ b/api/core/model_runtime/model_providers/siliconflow/llm/glm4-9b-chat.yaml @@ -21,6 +21,15 @@ parameter_rules: en_US: Specifies the upper limit on the length of generated results. If the generated results are truncated, you can increase this parameter. - name: top_p use_template: top_p + - name: top_k + label: + zh_Hans: 取样数量 + en_US: Top k + type: int + help: + zh_Hans: 仅从每个后续标记的前 K 个选项中采样。 + en_US: Only sample from the top K options for each subsequent token. + required: false - name: frequency_penalty use_template: frequency_penalty pricing: diff --git a/api/core/model_runtime/model_providers/siliconflow/llm/internlm2_5-20b-chat.yaml b/api/core/model_runtime/model_providers/siliconflow/llm/internlm2_5-20b-chat.yaml index d9663582e..60157c2b4 100644 --- a/api/core/model_runtime/model_providers/siliconflow/llm/internlm2_5-20b-chat.yaml +++ b/api/core/model_runtime/model_providers/siliconflow/llm/internlm2_5-20b-chat.yaml @@ -21,6 +21,15 @@ parameter_rules: en_US: Specifies the upper limit on the length of generated results. If the generated results are truncated, you can increase this parameter. - name: top_p use_template: top_p + - name: top_k + label: + zh_Hans: 取样数量 + en_US: Top k + type: int + help: + zh_Hans: 仅从每个后续标记的前 K 个选项中采样。 + en_US: Only sample from the top K options for each subsequent token. + required: false - name: frequency_penalty use_template: frequency_penalty pricing: diff --git a/api/core/model_runtime/model_providers/siliconflow/llm/internlm2_5-7b-chat.yaml b/api/core/model_runtime/model_providers/siliconflow/llm/internlm2_5-7b-chat.yaml index 73ad4480a..faf4af7ea 100644 --- a/api/core/model_runtime/model_providers/siliconflow/llm/internlm2_5-7b-chat.yaml +++ b/api/core/model_runtime/model_providers/siliconflow/llm/internlm2_5-7b-chat.yaml @@ -21,6 +21,15 @@ parameter_rules: en_US: Specifies the upper limit on the length of generated results. If the generated results are truncated, you can increase this parameter. - name: top_p use_template: top_p + - name: top_k + label: + zh_Hans: 取样数量 + en_US: Top k + type: int + help: + zh_Hans: 仅从每个后续标记的前 K 个选项中采样。 + en_US: Only sample from the top K options for each subsequent token. + required: false - name: frequency_penalty use_template: frequency_penalty pricing: diff --git a/api/core/model_runtime/model_providers/siliconflow/llm/meta-mlama-3-70b-instruct.yaml b/api/core/model_runtime/model_providers/siliconflow/llm/meta-mlama-3-70b-instruct.yaml index 9993d781a..d01770cb0 100644 --- a/api/core/model_runtime/model_providers/siliconflow/llm/meta-mlama-3-70b-instruct.yaml +++ b/api/core/model_runtime/model_providers/siliconflow/llm/meta-mlama-3-70b-instruct.yaml @@ -21,6 +21,15 @@ parameter_rules: en_US: Specifies the upper limit on the length of generated results. If the generated results are truncated, you can increase this parameter. - name: top_p use_template: top_p + - name: top_k + label: + zh_Hans: 取样数量 + en_US: Top k + type: int + help: + zh_Hans: 仅从每个后续标记的前 K 个选项中采样。 + en_US: Only sample from the top K options for each subsequent token. + required: false - name: frequency_penalty use_template: frequency_penalty pricing: diff --git a/api/core/model_runtime/model_providers/siliconflow/llm/meta-mlama-3-8b-instruct.yaml b/api/core/model_runtime/model_providers/siliconflow/llm/meta-mlama-3-8b-instruct.yaml index 60e376478..3cd75d89e 100644 --- a/api/core/model_runtime/model_providers/siliconflow/llm/meta-mlama-3-8b-instruct.yaml +++ b/api/core/model_runtime/model_providers/siliconflow/llm/meta-mlama-3-8b-instruct.yaml @@ -21,6 +21,15 @@ parameter_rules: en_US: Specifies the upper limit on the length of generated results. If the generated results are truncated, you can increase this parameter. - name: top_p use_template: top_p + - name: top_k + label: + zh_Hans: 取样数量 + en_US: Top k + type: int + help: + zh_Hans: 仅从每个后续标记的前 K 个选项中采样。 + en_US: Only sample from the top K options for each subsequent token. + required: false - name: frequency_penalty use_template: frequency_penalty pricing: diff --git a/api/core/model_runtime/model_providers/siliconflow/llm/meta-mlama-3.1-405b-instruct.yaml b/api/core/model_runtime/model_providers/siliconflow/llm/meta-mlama-3.1-405b-instruct.yaml index f992660aa..3506a70bc 100644 --- a/api/core/model_runtime/model_providers/siliconflow/llm/meta-mlama-3.1-405b-instruct.yaml +++ b/api/core/model_runtime/model_providers/siliconflow/llm/meta-mlama-3.1-405b-instruct.yaml @@ -21,6 +21,15 @@ parameter_rules: en_US: Specifies the upper limit on the length of generated results. If the generated results are truncated, you can increase this parameter. - name: top_p use_template: top_p + - name: top_k + label: + zh_Hans: 取样数量 + en_US: Top k + type: int + help: + zh_Hans: 仅从每个后续标记的前 K 个选项中采样。 + en_US: Only sample from the top K options for each subsequent token. + required: false - name: frequency_penalty use_template: frequency_penalty pricing: diff --git a/api/core/model_runtime/model_providers/siliconflow/llm/meta-mlama-3.1-70b-instruct.yaml b/api/core/model_runtime/model_providers/siliconflow/llm/meta-mlama-3.1-70b-instruct.yaml index 1c69d63a4..994a754a8 100644 --- a/api/core/model_runtime/model_providers/siliconflow/llm/meta-mlama-3.1-70b-instruct.yaml +++ b/api/core/model_runtime/model_providers/siliconflow/llm/meta-mlama-3.1-70b-instruct.yaml @@ -21,6 +21,15 @@ parameter_rules: en_US: Specifies the upper limit on the length of generated results. If the generated results are truncated, you can increase this parameter. - name: top_p use_template: top_p + - name: top_k + label: + zh_Hans: 取样数量 + en_US: Top k + type: int + help: + zh_Hans: 仅从每个后续标记的前 K 个选项中采样。 + en_US: Only sample from the top K options for each subsequent token. + required: false - name: frequency_penalty use_template: frequency_penalty pricing: diff --git a/api/core/model_runtime/model_providers/siliconflow/llm/meta-mlama-3.1-8b-instruct.yaml b/api/core/model_runtime/model_providers/siliconflow/llm/meta-mlama-3.1-8b-instruct.yaml index a97002a5c..ebfa9aac9 100644 --- a/api/core/model_runtime/model_providers/siliconflow/llm/meta-mlama-3.1-8b-instruct.yaml +++ b/api/core/model_runtime/model_providers/siliconflow/llm/meta-mlama-3.1-8b-instruct.yaml @@ -21,6 +21,15 @@ parameter_rules: en_US: Specifies the upper limit on the length of generated results. If the generated results are truncated, you can increase this parameter. - name: top_p use_template: top_p + - name: top_k + label: + zh_Hans: 取样数量 + en_US: Top k + type: int + help: + zh_Hans: 仅从每个后续标记的前 K 个选项中采样。 + en_US: Only sample from the top K options for each subsequent token. + required: false - name: frequency_penalty use_template: frequency_penalty pricing: diff --git a/api/core/model_runtime/model_providers/siliconflow/llm/mistral-7b-instruct-v0.2.yaml b/api/core/model_runtime/model_providers/siliconflow/llm/mistral-7b-instruct-v0.2.yaml index 89fb153ba..a71d8688a 100644 --- a/api/core/model_runtime/model_providers/siliconflow/llm/mistral-7b-instruct-v0.2.yaml +++ b/api/core/model_runtime/model_providers/siliconflow/llm/mistral-7b-instruct-v0.2.yaml @@ -21,6 +21,15 @@ parameter_rules: en_US: Specifies the upper limit on the length of generated results. If the generated results are truncated, you can increase this parameter. - name: top_p use_template: top_p + - name: top_k + label: + zh_Hans: 取样数量 + en_US: Top k + type: int + help: + zh_Hans: 仅从每个后续标记的前 K 个选项中采样。 + en_US: Only sample from the top K options for each subsequent token. + required: false - name: frequency_penalty use_template: frequency_penalty pricing: diff --git a/api/core/model_runtime/model_providers/siliconflow/llm/mistral-8x7b-instruct-v0.1.yaml b/api/core/model_runtime/model_providers/siliconflow/llm/mistral-8x7b-instruct-v0.1.yaml index 2785e7496..db45a75c6 100644 --- a/api/core/model_runtime/model_providers/siliconflow/llm/mistral-8x7b-instruct-v0.1.yaml +++ b/api/core/model_runtime/model_providers/siliconflow/llm/mistral-8x7b-instruct-v0.1.yaml @@ -21,6 +21,15 @@ parameter_rules: en_US: Specifies the upper limit on the length of generated results. If the generated results are truncated, you can increase this parameter. - name: top_p use_template: top_p + - name: top_k + label: + zh_Hans: 取样数量 + en_US: Top k + type: int + help: + zh_Hans: 仅从每个后续标记的前 K 个选项中采样。 + en_US: Only sample from the top K options for each subsequent token. + required: false - name: frequency_penalty use_template: frequency_penalty pricing: diff --git a/api/core/model_runtime/model_providers/siliconflow/llm/qwen2-1.5b-instruct.yaml b/api/core/model_runtime/model_providers/siliconflow/llm/qwen2-1.5b-instruct.yaml index f6c976af8..bec5d37c5 100644 --- a/api/core/model_runtime/model_providers/siliconflow/llm/qwen2-1.5b-instruct.yaml +++ b/api/core/model_runtime/model_providers/siliconflow/llm/qwen2-1.5b-instruct.yaml @@ -21,6 +21,15 @@ parameter_rules: en_US: Specifies the upper limit on the length of generated results. If the generated results are truncated, you can increase this parameter. - name: top_p use_template: top_p + - name: top_k + label: + zh_Hans: 取样数量 + en_US: Top k + type: int + help: + zh_Hans: 仅从每个后续标记的前 K 个选项中采样。 + en_US: Only sample from the top K options for each subsequent token. + required: false - name: frequency_penalty use_template: frequency_penalty pricing: diff --git a/api/core/model_runtime/model_providers/siliconflow/llm/qwen2-57b-a14b-instruct.yaml b/api/core/model_runtime/model_providers/siliconflow/llm/qwen2-57b-a14b-instruct.yaml index a996e919e..b2461335f 100644 --- a/api/core/model_runtime/model_providers/siliconflow/llm/qwen2-57b-a14b-instruct.yaml +++ b/api/core/model_runtime/model_providers/siliconflow/llm/qwen2-57b-a14b-instruct.yaml @@ -21,6 +21,15 @@ parameter_rules: en_US: Specifies the upper limit on the length of generated results. If the generated results are truncated, you can increase this parameter. - name: top_p use_template: top_p + - name: top_k + label: + zh_Hans: 取样数量 + en_US: Top k + type: int + help: + zh_Hans: 仅从每个后续标记的前 K 个选项中采样。 + en_US: Only sample from the top K options for each subsequent token. + required: false - name: frequency_penalty use_template: frequency_penalty pricing: diff --git a/api/core/model_runtime/model_providers/siliconflow/llm/qwen2-72b-instruct.yaml b/api/core/model_runtime/model_providers/siliconflow/llm/qwen2-72b-instruct.yaml index a6e2c22da..e0f23bd89 100644 --- a/api/core/model_runtime/model_providers/siliconflow/llm/qwen2-72b-instruct.yaml +++ b/api/core/model_runtime/model_providers/siliconflow/llm/qwen2-72b-instruct.yaml @@ -21,6 +21,15 @@ parameter_rules: en_US: Specifies the upper limit on the length of generated results. If the generated results are truncated, you can increase this parameter. - name: top_p use_template: top_p + - name: top_k + label: + zh_Hans: 取样数量 + en_US: Top k + type: int + help: + zh_Hans: 仅从每个后续标记的前 K 个选项中采样。 + en_US: Only sample from the top K options for each subsequent token. + required: false - name: frequency_penalty use_template: frequency_penalty pricing: diff --git a/api/core/model_runtime/model_providers/siliconflow/llm/qwen2-7b-instruct.yaml b/api/core/model_runtime/model_providers/siliconflow/llm/qwen2-7b-instruct.yaml index d8bea5e12..47a9da811 100644 --- a/api/core/model_runtime/model_providers/siliconflow/llm/qwen2-7b-instruct.yaml +++ b/api/core/model_runtime/model_providers/siliconflow/llm/qwen2-7b-instruct.yaml @@ -21,6 +21,15 @@ parameter_rules: en_US: Specifies the upper limit on the length of generated results. If the generated results are truncated, you can increase this parameter. - name: top_p use_template: top_p + - name: top_k + label: + zh_Hans: 取样数量 + en_US: Top k + type: int + help: + zh_Hans: 仅从每个后续标记的前 K 个选项中采样。 + en_US: Only sample from the top K options for each subsequent token. + required: false - name: frequency_penalty use_template: frequency_penalty pricing: diff --git a/api/core/model_runtime/model_providers/siliconflow/llm/qwen2.5-14b-instruct.yaml b/api/core/model_runtime/model_providers/siliconflow/llm/qwen2.5-14b-instruct.yaml index 02a401464..9cc5ac4c9 100644 --- a/api/core/model_runtime/model_providers/siliconflow/llm/qwen2.5-14b-instruct.yaml +++ b/api/core/model_runtime/model_providers/siliconflow/llm/qwen2.5-14b-instruct.yaml @@ -21,6 +21,15 @@ parameter_rules: en_US: Specifies the upper limit on the length of generated results. If the generated results are truncated, you can increase this parameter. - name: top_p use_template: top_p + - name: top_k + label: + zh_Hans: 取样数量 + en_US: Top k + type: int + help: + zh_Hans: 仅从每个后续标记的前 K 个选项中采样。 + en_US: Only sample from the top K options for each subsequent token. + required: false - name: frequency_penalty use_template: frequency_penalty pricing: diff --git a/api/core/model_runtime/model_providers/siliconflow/llm/qwen2.5-32b-instruct.yaml b/api/core/model_runtime/model_providers/siliconflow/llm/qwen2.5-32b-instruct.yaml index d084617e7..c7fb21e9e 100644 --- a/api/core/model_runtime/model_providers/siliconflow/llm/qwen2.5-32b-instruct.yaml +++ b/api/core/model_runtime/model_providers/siliconflow/llm/qwen2.5-32b-instruct.yaml @@ -21,6 +21,15 @@ parameter_rules: en_US: Specifies the upper limit on the length of generated results. If the generated results are truncated, you can increase this parameter. - name: top_p use_template: top_p + - name: top_k + label: + zh_Hans: 取样数量 + en_US: Top k + type: int + help: + zh_Hans: 仅从每个后续标记的前 K 个选项中采样。 + en_US: Only sample from the top K options for each subsequent token. + required: false - name: frequency_penalty use_template: frequency_penalty pricing: diff --git a/api/core/model_runtime/model_providers/siliconflow/llm/qwen2.5-72b-instruct.yaml b/api/core/model_runtime/model_providers/siliconflow/llm/qwen2.5-72b-instruct.yaml index dfbad2494..03136c88a 100644 --- a/api/core/model_runtime/model_providers/siliconflow/llm/qwen2.5-72b-instruct.yaml +++ b/api/core/model_runtime/model_providers/siliconflow/llm/qwen2.5-72b-instruct.yaml @@ -21,6 +21,15 @@ parameter_rules: en_US: Specifies the upper limit on the length of generated results. If the generated results are truncated, you can increase this parameter. - name: top_p use_template: top_p + - name: top_k + label: + zh_Hans: 取样数量 + en_US: Top k + type: int + help: + zh_Hans: 仅从每个后续标记的前 K 个选项中采样。 + en_US: Only sample from the top K options for each subsequent token. + required: false - name: frequency_penalty use_template: frequency_penalty pricing: diff --git a/api/core/model_runtime/model_providers/siliconflow/llm/qwen2.5-7b-instruct.yaml b/api/core/model_runtime/model_providers/siliconflow/llm/qwen2.5-7b-instruct.yaml index cdc8ffc4d..99412adde 100644 --- a/api/core/model_runtime/model_providers/siliconflow/llm/qwen2.5-7b-instruct.yaml +++ b/api/core/model_runtime/model_providers/siliconflow/llm/qwen2.5-7b-instruct.yaml @@ -21,6 +21,15 @@ parameter_rules: en_US: Specifies the upper limit on the length of generated results. If the generated results are truncated, you can increase this parameter. - name: top_p use_template: top_p + - name: top_k + label: + zh_Hans: 取样数量 + en_US: Top k + type: int + help: + zh_Hans: 仅从每个后续标记的前 K 个选项中采样。 + en_US: Only sample from the top K options for each subsequent token. + required: false - name: frequency_penalty use_template: frequency_penalty pricing: diff --git a/api/core/model_runtime/model_providers/siliconflow/llm/yi-1.5-34b-chat.yaml b/api/core/model_runtime/model_providers/siliconflow/llm/yi-1.5-34b-chat.yaml index 864ba46f1..3e25f8236 100644 --- a/api/core/model_runtime/model_providers/siliconflow/llm/yi-1.5-34b-chat.yaml +++ b/api/core/model_runtime/model_providers/siliconflow/llm/yi-1.5-34b-chat.yaml @@ -21,6 +21,15 @@ parameter_rules: en_US: Specifies the upper limit on the length of generated results. If the generated results are truncated, you can increase this parameter. - name: top_p use_template: top_p + - name: top_k + label: + zh_Hans: 取样数量 + en_US: Top k + type: int + help: + zh_Hans: 仅从每个后续标记的前 K 个选项中采样。 + en_US: Only sample from the top K options for each subsequent token. + required: false - name: frequency_penalty use_template: frequency_penalty pricing: diff --git a/api/core/model_runtime/model_providers/siliconflow/llm/yi-1.5-6b-chat.yaml b/api/core/model_runtime/model_providers/siliconflow/llm/yi-1.5-6b-chat.yaml index fe4c8b4b3..827b2ce1e 100644 --- a/api/core/model_runtime/model_providers/siliconflow/llm/yi-1.5-6b-chat.yaml +++ b/api/core/model_runtime/model_providers/siliconflow/llm/yi-1.5-6b-chat.yaml @@ -21,6 +21,15 @@ parameter_rules: en_US: Specifies the upper limit on the length of generated results. If the generated results are truncated, you can increase this parameter. - name: top_p use_template: top_p + - name: top_k + label: + zh_Hans: 取样数量 + en_US: Top k + type: int + help: + zh_Hans: 仅从每个后续标记的前 K 个选项中采样。 + en_US: Only sample from the top K options for each subsequent token. + required: false - name: frequency_penalty use_template: frequency_penalty pricing: diff --git a/api/core/model_runtime/model_providers/siliconflow/llm/yi-1.5-9b-chat.yaml b/api/core/model_runtime/model_providers/siliconflow/llm/yi-1.5-9b-chat.yaml index c61f0dc53..112fcbfe9 100644 --- a/api/core/model_runtime/model_providers/siliconflow/llm/yi-1.5-9b-chat.yaml +++ b/api/core/model_runtime/model_providers/siliconflow/llm/yi-1.5-9b-chat.yaml @@ -21,6 +21,15 @@ parameter_rules: en_US: Specifies the upper limit on the length of generated results. If the generated results are truncated, you can increase this parameter. - name: top_p use_template: top_p + - name: top_k + label: + zh_Hans: 取样数量 + en_US: Top k + type: int + help: + zh_Hans: 仅从每个后续标记的前 K 个选项中采样。 + en_US: Only sample from the top K options for each subsequent token. + required: false - name: frequency_penalty use_template: frequency_penalty pricing: