diff --git a/src/content/workers-ai-models/deepseek-coder-6.7b-base-awq.json b/src/content/workers-ai-models/deepseek-coder-6.7b-base-awq.json index 0c150623e3b5115..420bef8ac689167 100644 --- a/src/content/workers-ai-models/deepseek-coder-6.7b-base-awq.json +++ b/src/content/workers-ai-models/deepseek-coder-6.7b-base-awq.json @@ -78,8 +78,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -102,13 +102,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } @@ -338,8 +338,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -362,13 +362,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } diff --git a/src/content/workers-ai-models/deepseek-coder-6.7b-instruct-awq.json b/src/content/workers-ai-models/deepseek-coder-6.7b-instruct-awq.json index 372a20033cf69cb..37754a5d1079e06 100644 --- a/src/content/workers-ai-models/deepseek-coder-6.7b-instruct-awq.json +++ b/src/content/workers-ai-models/deepseek-coder-6.7b-instruct-awq.json @@ -78,8 +78,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -102,13 +102,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } @@ -338,8 +338,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -362,13 +362,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } diff --git a/src/content/workers-ai-models/deepseek-math-7b-instruct.json b/src/content/workers-ai-models/deepseek-math-7b-instruct.json index c0fd00abd381479..a841324071e880b 100644 --- a/src/content/workers-ai-models/deepseek-math-7b-instruct.json +++ b/src/content/workers-ai-models/deepseek-math-7b-instruct.json @@ -82,8 +82,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -106,13 +106,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } @@ -342,8 +342,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -366,13 +366,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } diff --git a/src/content/workers-ai-models/deepseek-r1-distill-qwen-32b.json b/src/content/workers-ai-models/deepseek-r1-distill-qwen-32b.json index 18959db015dcd3a..5ad6730fa306c21 100644 --- a/src/content/workers-ai-models/deepseek-r1-distill-qwen-32b.json +++ b/src/content/workers-ai-models/deepseek-r1-distill-qwen-32b.json @@ -89,8 +89,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -113,13 +113,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } @@ -349,8 +349,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -373,13 +373,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } diff --git a/src/content/workers-ai-models/discolm-german-7b-v1-awq.json b/src/content/workers-ai-models/discolm-german-7b-v1-awq.json index dd68a3ed998410f..3cc9a4bf2518ef3 100644 --- a/src/content/workers-ai-models/discolm-german-7b-v1-awq.json +++ b/src/content/workers-ai-models/discolm-german-7b-v1-awq.json @@ -78,8 +78,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -102,13 +102,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } @@ -338,8 +338,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -362,13 +362,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } diff --git a/src/content/workers-ai-models/gemma-2b-it-lora.json b/src/content/workers-ai-models/gemma-2b-it-lora.json index 3a4757b96af73fc..05ea356da537f28 100644 --- a/src/content/workers-ai-models/gemma-2b-it-lora.json +++ b/src/content/workers-ai-models/gemma-2b-it-lora.json @@ -78,8 +78,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -102,13 +102,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } @@ -338,8 +338,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -362,13 +362,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } diff --git a/src/content/workers-ai-models/gemma-7b-it.json b/src/content/workers-ai-models/gemma-7b-it.json index 42dec8cda5d5de0..2ed52e6d7af0c68 100644 --- a/src/content/workers-ai-models/gemma-7b-it.json +++ b/src/content/workers-ai-models/gemma-7b-it.json @@ -86,8 +86,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -110,13 +110,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } @@ -346,8 +346,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -370,13 +370,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } diff --git a/src/content/workers-ai-models/llama-2-13b-chat-awq.json b/src/content/workers-ai-models/llama-2-13b-chat-awq.json index d2a9117a30a4a33..bc87f796b423921 100644 --- a/src/content/workers-ai-models/llama-2-13b-chat-awq.json +++ b/src/content/workers-ai-models/llama-2-13b-chat-awq.json @@ -78,8 +78,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -102,13 +102,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } @@ -338,8 +338,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -362,13 +362,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } diff --git a/src/content/workers-ai-models/llama-2-7b-chat-fp16.json b/src/content/workers-ai-models/llama-2-7b-chat-fp16.json index 41d3cb0d7d17991..a295e54f53a3d2b 100644 --- a/src/content/workers-ai-models/llama-2-7b-chat-fp16.json +++ b/src/content/workers-ai-models/llama-2-7b-chat-fp16.json @@ -93,8 +93,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -117,13 +117,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } @@ -353,8 +353,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -377,13 +377,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } diff --git a/src/content/workers-ai-models/llama-2-7b-chat-hf-lora.json b/src/content/workers-ai-models/llama-2-7b-chat-hf-lora.json index 8c27b2d12da2261..4f3f6ed433c9bc7 100644 --- a/src/content/workers-ai-models/llama-2-7b-chat-hf-lora.json +++ b/src/content/workers-ai-models/llama-2-7b-chat-hf-lora.json @@ -78,8 +78,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -102,13 +102,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } @@ -338,8 +338,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -362,13 +362,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } diff --git a/src/content/workers-ai-models/llama-2-7b-chat-int8.json b/src/content/workers-ai-models/llama-2-7b-chat-int8.json index b9f1d1715c1f585..b0e8daacb21d54a 100644 --- a/src/content/workers-ai-models/llama-2-7b-chat-int8.json +++ b/src/content/workers-ai-models/llama-2-7b-chat-int8.json @@ -70,8 +70,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -94,13 +94,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } @@ -330,8 +330,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -354,13 +354,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } diff --git a/src/content/workers-ai-models/llama-3-8b-instruct-awq.json b/src/content/workers-ai-models/llama-3-8b-instruct-awq.json index a8f0d7a7edc5b0b..2c91f4573d0214d 100644 --- a/src/content/workers-ai-models/llama-3-8b-instruct-awq.json +++ b/src/content/workers-ai-models/llama-3-8b-instruct-awq.json @@ -93,8 +93,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -117,13 +117,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } @@ -353,8 +353,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -377,13 +377,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } diff --git a/src/content/workers-ai-models/llama-3-8b-instruct.json b/src/content/workers-ai-models/llama-3-8b-instruct.json index 333986179cd91e7..f02bf99c2bb11e2 100644 --- a/src/content/workers-ai-models/llama-3-8b-instruct.json +++ b/src/content/workers-ai-models/llama-3-8b-instruct.json @@ -93,8 +93,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -117,13 +117,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } @@ -353,8 +353,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -377,13 +377,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } diff --git a/src/content/workers-ai-models/llama-3.1-8b-instruct-awq.json b/src/content/workers-ai-models/llama-3.1-8b-instruct-awq.json index b88b86b10d43479..99eb067f76644b0 100644 --- a/src/content/workers-ai-models/llama-3.1-8b-instruct-awq.json +++ b/src/content/workers-ai-models/llama-3.1-8b-instruct-awq.json @@ -89,8 +89,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -113,13 +113,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } @@ -349,8 +349,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -373,13 +373,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } diff --git a/src/content/workers-ai-models/llama-3.2-1b-instruct.json b/src/content/workers-ai-models/llama-3.2-1b-instruct.json index e394312e3d4edfb..cc37d4210fcdaf4 100644 --- a/src/content/workers-ai-models/llama-3.2-1b-instruct.json +++ b/src/content/workers-ai-models/llama-3.2-1b-instruct.json @@ -89,8 +89,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -113,13 +113,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } @@ -349,8 +349,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -373,13 +373,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } diff --git a/src/content/workers-ai-models/llama-3.2-3b-instruct.json b/src/content/workers-ai-models/llama-3.2-3b-instruct.json index f6022bfa88aca52..d393afb1cda371c 100644 --- a/src/content/workers-ai-models/llama-3.2-3b-instruct.json +++ b/src/content/workers-ai-models/llama-3.2-3b-instruct.json @@ -89,8 +89,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -113,13 +113,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } @@ -349,8 +349,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -373,13 +373,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } diff --git a/src/content/workers-ai-models/llama-3.3-70b-instruct-fp8-fast.json b/src/content/workers-ai-models/llama-3.3-70b-instruct-fp8-fast.json index 381cd92da9c6905..a035832c1735b5a 100644 --- a/src/content/workers-ai-models/llama-3.3-70b-instruct-fp8-fast.json +++ b/src/content/workers-ai-models/llama-3.3-70b-instruct-fp8-fast.json @@ -97,8 +97,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -121,13 +121,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } @@ -357,8 +357,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -381,13 +381,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } diff --git a/src/content/workers-ai-models/llama-4-scout-17b-16e-instruct.json b/src/content/workers-ai-models/llama-4-scout-17b-16e-instruct.json index b7935e245d93ea5..9211d117f1c6dee 100644 --- a/src/content/workers-ai-models/llama-4-scout-17b-16e-instruct.json +++ b/src/content/workers-ai-models/llama-4-scout-17b-16e-instruct.json @@ -11,6 +11,10 @@ "created_at": "2025-04-05 20:25:56.137", "tags": [], "properties": [ + { + "property_id": "async_queue", + "value": "true" + }, { "property_id": "context_window", "value": "131000" diff --git a/src/content/workers-ai-models/llamaguard-7b-awq.json b/src/content/workers-ai-models/llamaguard-7b-awq.json index c1a23a4aa1be01c..ef3f7e6d9385a32 100644 --- a/src/content/workers-ai-models/llamaguard-7b-awq.json +++ b/src/content/workers-ai-models/llamaguard-7b-awq.json @@ -74,8 +74,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -98,13 +98,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } @@ -334,8 +334,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -358,13 +358,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } diff --git a/src/content/workers-ai-models/mistral-7b-instruct-v0.1-awq.json b/src/content/workers-ai-models/mistral-7b-instruct-v0.1-awq.json index cebb5e6c79f05b7..dff17a736f1e7e2 100644 --- a/src/content/workers-ai-models/mistral-7b-instruct-v0.1-awq.json +++ b/src/content/workers-ai-models/mistral-7b-instruct-v0.1-awq.json @@ -78,8 +78,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -102,13 +102,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } @@ -338,8 +338,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -362,13 +362,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } diff --git a/src/content/workers-ai-models/mistral-7b-instruct-v0.1.json b/src/content/workers-ai-models/mistral-7b-instruct-v0.1.json index afe429f4438c368..8b204412036b005 100644 --- a/src/content/workers-ai-models/mistral-7b-instruct-v0.1.json +++ b/src/content/workers-ai-models/mistral-7b-instruct-v0.1.json @@ -93,8 +93,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -117,13 +117,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } @@ -353,8 +353,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -377,13 +377,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } diff --git a/src/content/workers-ai-models/mistral-7b-instruct-v0.2-lora.json b/src/content/workers-ai-models/mistral-7b-instruct-v0.2-lora.json index c886d9508393904..05b324ef2042725 100644 --- a/src/content/workers-ai-models/mistral-7b-instruct-v0.2-lora.json +++ b/src/content/workers-ai-models/mistral-7b-instruct-v0.2-lora.json @@ -78,8 +78,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -102,13 +102,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } @@ -338,8 +338,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -362,13 +362,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } diff --git a/src/content/workers-ai-models/mistral-7b-instruct-v0.2.json b/src/content/workers-ai-models/mistral-7b-instruct-v0.2.json index 379eddd42728839..acc2f504c59557f 100644 --- a/src/content/workers-ai-models/mistral-7b-instruct-v0.2.json +++ b/src/content/workers-ai-models/mistral-7b-instruct-v0.2.json @@ -94,8 +94,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -118,13 +118,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } @@ -354,8 +354,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -378,13 +378,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } diff --git a/src/content/workers-ai-models/openchat-3.5-0106.json b/src/content/workers-ai-models/openchat-3.5-0106.json index e1edd0767514d5d..d2308b29abdd324 100644 --- a/src/content/workers-ai-models/openchat-3.5-0106.json +++ b/src/content/workers-ai-models/openchat-3.5-0106.json @@ -78,8 +78,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -102,13 +102,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } @@ -338,8 +338,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -362,13 +362,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } diff --git a/src/content/workers-ai-models/openhermes-2.5-mistral-7b-awq.json b/src/content/workers-ai-models/openhermes-2.5-mistral-7b-awq.json index 133cf985ebfc36c..c679f11edf5f730 100644 --- a/src/content/workers-ai-models/openhermes-2.5-mistral-7b-awq.json +++ b/src/content/workers-ai-models/openhermes-2.5-mistral-7b-awq.json @@ -74,8 +74,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -98,13 +98,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } @@ -334,8 +334,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -358,13 +358,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } diff --git a/src/content/workers-ai-models/phi-2.json b/src/content/workers-ai-models/phi-2.json index 64eaa4b7f9bc8c4..86eb905fe92d6ec 100644 --- a/src/content/workers-ai-models/phi-2.json +++ b/src/content/workers-ai-models/phi-2.json @@ -78,8 +78,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -102,13 +102,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } @@ -338,8 +338,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -362,13 +362,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } diff --git a/src/content/workers-ai-models/qwen1.5-0.5b-chat.json b/src/content/workers-ai-models/qwen1.5-0.5b-chat.json index aa0d134db9d561e..785da210b569376 100644 --- a/src/content/workers-ai-models/qwen1.5-0.5b-chat.json +++ b/src/content/workers-ai-models/qwen1.5-0.5b-chat.json @@ -78,8 +78,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -102,13 +102,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } @@ -338,8 +338,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -362,13 +362,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } diff --git a/src/content/workers-ai-models/qwen1.5-1.8b-chat.json b/src/content/workers-ai-models/qwen1.5-1.8b-chat.json index 4e66016ae192943..9f980d704dfa2ca 100644 --- a/src/content/workers-ai-models/qwen1.5-1.8b-chat.json +++ b/src/content/workers-ai-models/qwen1.5-1.8b-chat.json @@ -78,8 +78,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -102,13 +102,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } @@ -338,8 +338,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -362,13 +362,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } diff --git a/src/content/workers-ai-models/qwen1.5-14b-chat-awq.json b/src/content/workers-ai-models/qwen1.5-14b-chat-awq.json index 5f3650c3fdadb77..f8d47c54777e87f 100644 --- a/src/content/workers-ai-models/qwen1.5-14b-chat-awq.json +++ b/src/content/workers-ai-models/qwen1.5-14b-chat-awq.json @@ -78,8 +78,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -102,13 +102,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } @@ -338,8 +338,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -362,13 +362,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } diff --git a/src/content/workers-ai-models/qwen1.5-7b-chat-awq.json b/src/content/workers-ai-models/qwen1.5-7b-chat-awq.json index 442115def594cfc..1ce2a9adc0e4ece 100644 --- a/src/content/workers-ai-models/qwen1.5-7b-chat-awq.json +++ b/src/content/workers-ai-models/qwen1.5-7b-chat-awq.json @@ -78,8 +78,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -102,13 +102,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } @@ -338,8 +338,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -362,13 +362,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } diff --git a/src/content/workers-ai-models/sqlcoder-7b-2.json b/src/content/workers-ai-models/sqlcoder-7b-2.json index c8b0139523f5e3c..b3ac9c706f9d6d4 100644 --- a/src/content/workers-ai-models/sqlcoder-7b-2.json +++ b/src/content/workers-ai-models/sqlcoder-7b-2.json @@ -82,8 +82,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -106,13 +106,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } @@ -342,8 +342,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -366,13 +366,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } diff --git a/src/content/workers-ai-models/tinyllama-1.1b-chat-v1.0.json b/src/content/workers-ai-models/tinyllama-1.1b-chat-v1.0.json index 8c94700baa1181d..e15448674080ac4 100644 --- a/src/content/workers-ai-models/tinyllama-1.1b-chat-v1.0.json +++ b/src/content/workers-ai-models/tinyllama-1.1b-chat-v1.0.json @@ -78,8 +78,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -102,13 +102,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } @@ -338,8 +338,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -362,13 +362,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } diff --git a/src/content/workers-ai-models/zephyr-7b-beta-awq.json b/src/content/workers-ai-models/zephyr-7b-beta-awq.json index ca781d97c42204d..c8a428c43bbd1c9 100644 --- a/src/content/workers-ai-models/zephyr-7b-beta-awq.json +++ b/src/content/workers-ai-models/zephyr-7b-beta-awq.json @@ -78,8 +78,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -102,13 +102,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." } @@ -338,8 +338,8 @@ }, "top_p": { "type": "number", - "minimum": 0, - "maximum": 2, + "minimum": 0.001, + "maximum": 1, "description": "Adjusts the creativity of the AI's responses by controlling how many possible words it considers. Lower values make outputs more predictable; higher values allow for more varied and creative responses." }, "top_k": { @@ -362,13 +362,13 @@ }, "frequency_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Decreases the likelihood of the model repeating the same lines verbatim." }, "presence_penalty": { "type": "number", - "minimum": 0, + "minimum": -2, "maximum": 2, "description": "Increases the likelihood of the model introducing new topics." }