diff --git a/models.yaml b/models.yaml index 5261b8da..70568178 100644 --- a/models.yaml +++ b/models.yaml @@ -87,6 +87,8 @@ - name: gemini-1.5-pro-exp-0827 max_input_tokens: 2097152 max_output_tokens: 8192 + input_price: 0 + output_price: 0 supports_vision: true supports_function_calling: true - name: gemini-1.5-flash-latest @@ -103,7 +105,7 @@ output_price: 0 supports_vision: true supports_function_calling: true - - name: gemini-1.5-flash-8b-exp-0827 + - name: gemini-1.5-flash-8b-exp-0924 max_input_tokens: 1048576 max_output_tokens: 8192 input_price: 0 @@ -163,25 +165,30 @@ supports_function_calling: true # Links: -# - https://docs.mistral.ai/getting-started/models/ +# - https://docs.mistral.ai/getting-started/models/models_overview/ # - https://mistral.ai/technology/#pricing # - https://docs.mistral.ai/api/ - platform: mistral models: - name: mistral-large-latest max_input_tokens: 128000 - input_price: 3 - output_price: 9 + input_price: 2 + output_price: 6 supports_function_calling: true - - name: open-mistral-nemo - max_input_tokens: 128000 - input_price: 0.3 - output_price: 0.3 + - name: mistral-small-latest + max_input_tokens: 32000 + input_price: 0.2 + output_price: 0.6 supports_function_calling: true - name: codestral-latest max_input_tokens: 32000 - input_price: 1 - output_price: 3 + input_price: 0.2 + output_price: 0.6 + - name: open-mistral-nemo + max_input_tokens: 128000 + input_price: 0.15 + output_price: 0.15 + supports_function_calling: true - name: open-codestral-mamba max_input_tokens: 256000 input_price: 0.25 @@ -221,23 +228,27 @@ # - https://docs.cohere.com/reference/chat - platform: cohere models: - - name: command-r-plus + - name: command-r-plus-08-2024 max_input_tokens: 128000 + max_output_tokens: 4096 input_price: 2.5 output_price: 10 supports_function_calling: true - - name: command-r-plus-08-2024 + - name: command-r-plus max_input_tokens: 128000 + max_output_tokens: 4096 input_price: 2.5 output_price: 10 supports_function_calling: true - - name: command-r + - name: command-r-08-2024 max_input_tokens: 128000 + max_output_tokens: 4096 input_price: 0.15 output_price: 0.6 supports_function_calling: true - - name: command-r-08-2024 + - name: command-r max_input_tokens: 128000 + max_output_tokens: 4096 input_price: 0.15 output_price: 0.6 supports_function_calling: true @@ -324,10 +335,12 @@ max_input_tokens: 8192 input_price: 0 output_price: 0 + supports_function_calling: true - name: llama-3.1-8b-instant max_input_tokens: 8192 input_price: 0 output_price: 0 + supports_function_calling: true - name: gemma2-9b-it max_input_tokens: 8192 input_price: 0 @@ -344,17 +357,19 @@ supports_function_calling: true - name: gemma2 max_input_tokens: 8192 - - name: mistral-nemo + - name: qwen2.5 max_input_tokens: 128000 supports_function_calling: true - - name: mistral-large + - name: phi3.5 + max_input_tokens: 128000 + - name: mistral-small max_input_tokens: 128000 supports_function_calling: true - - name: deepseek-coder-v2 - max_input_tokens: 32768 - - name: phi3 + - name: mistral-nemo max_input_tokens: 128000 supports_function_calling: true + - name: deepseek-coder-v2 + max_input_tokens: 32768 - name: nomic-embed-text type: embedding max_tokens_per_chunk: 8192 @@ -368,18 +383,18 @@ # - https://cloud.google.com/vertex-ai/generative-ai/docs/model-reference/gemini - platform: vertexai models: - - name: gemini-1.5-pro-001 + - name: gemini-1.5-pro-002 max_input_tokens: 2097152 max_output_tokens: 8192 input_price: 1.25 output_price: 3.75 supports_vision: true supports_function_calling: true - - name: gemini-1.5-flash-001 - max_input_tokens: 1000000 + - name: gemini-1.5-flash-002 + max_input_tokens: 1048576 max_output_tokens: 8192 input_price: 0.01875 - output_price: 0.0375 + output_price: 0.075 supports_vision: true supports_function_calling: true - name: gemini-1.0-pro-002 @@ -422,18 +437,18 @@ supports_function_calling: true - name: mistral-large@2407 max_input_tokens: 128000 - input_price: 3 - output_price: 9 + input_price: 2 + output_price: 6 supports_function_calling: true - name: mistral-nemo@2407 max_input_tokens: 128000 - input_price: 0.3 - output_price: 0.3 + input_price: 0.15 + output_price: 0.15 supports_function_calling: true - name: codestral@2405 max_input_tokens: 32000 - input_price: 1 - output_price: 3 + input_price: 0.2 + output_price: 0.6 - name: text-embedding-004 type: embedding max_input_tokens: 20000 @@ -494,13 +509,13 @@ supports_function_calling: true - name: meta.llama3-1-70b-instruct-v1:0 max_input_tokens: 128000 - input_price: 2.65 - output_price: 3.5 + input_price: 0.99 + output_price: 0.99 supports_function_calling: true - name: meta.llama3-1-8b-instruct-v1:0 max_input_tokens: 128000 - input_price: 0.3 - output_price: 0.6 + input_price: 0.22 + output_price: 0.22 supports_function_calling: true - name: meta.llama3-70b-instruct-v1:0 max_input_tokens: 8192 @@ -512,8 +527,8 @@ output_price: 0.6 - name: mistral.mistral-large-2407-v1:0 max_input_tokens: 128000 - input_price: 3 - output_price: 9 + input_price: 2 + output_price: 6 supports_function_calling: true - name: cohere.command-r-plus-v1:0 max_input_tokens: 128000 @@ -537,6 +552,16 @@ max_tokens_per_chunk: 512 default_chunk_size: 1000 max_batch_size: 96 + - name: ai21.jamba-1-5-large-v1:0 + max_input_tokens: 256000 + input_price: 2 + output_price: 8 + supports_function_calling: true + - name: ai21.jamba-1-5-mini-v1:0 + max_input_tokens: 256000 + input_price: 0.2 + output_price: 0.4 + supports_function_calling: true # Links: # - https://developers.cloudflare.com/workers-ai/models/ @@ -630,6 +655,10 @@ input_price: 1.68 output_price: 1.68 supports_function_calling: true + - name: ernie-speed-pro-128k + max_input_tokens: 128000 + input_price: 0.056 + output_price: 0.112 - name: ernie-speed-128k max_input_tokens: 128000 input_price: 0 @@ -657,23 +686,27 @@ # - https://help.aliyun.com/zh/dashscope/developer-reference/use-qwen-by-api - platform: qianwen models: - - name: qwen-max - max_input_tokens: 8000 - input_price: 5.6 - output_price: 16.8 + - name: qwen-max-latest + max_input_tokens: 30720 + max_output_tokens: 8192 + input_price: 2.8 + output_price: 8.4 supports_function_calling: true - - name: qwen-max-longcontext - input_price: 5.6 - output_price: 16.8 - max_input_tokens: 30000 + - name: qwen-plus-latest + max_input_tokens: 128000 + max_output_tokens: 8192 + input_price: 0.112 + output_price: 0.28 supports_function_calling: true - - name: qwen-plus - max_input_tokens: 32000 - input_price: 0.56 - output_price: 1.68 + - name: qwen-turbo-latest + max_input_tokens: 129024 + max_output_tokens: 8192 + input_price: 0.042 + output_price: 0.084 supports_function_calling: true - - name: qwen-turbo - max_input_tokens: 8000 + - name: qwen-coder-turbo-latest + max_input_tokens: 129024 + max_output_tokens: 8192 input_price: 0.28 output_price: 0.84 supports_function_calling: true @@ -731,11 +764,7 @@ models: - name: deepseek-chat max_input_tokens: 32768 - input_price: 0.14 - output_price: 0.28 - supports_function_calling: true - - name: deepseek-coder - max_input_tokens: 32768 + max_output_tokens: 4096 input_price: 0.14 output_price: 0.28 supports_function_calling: true @@ -752,7 +781,7 @@ output_price: 7 supports_function_calling: true - name: glm-4-alltools - max_input_tokens: 2048 + max_input_tokens: 128000 input_price: 14 output_price: 14 supports_function_calling: true @@ -893,27 +922,27 @@ models: - name: meta-llama/Meta-Llama-3.1-405B-Instruct max_input_tokens: 32000 - input_price: 2.7 - output_price: 2.7 + input_price: 1.79 + output_price: 1.79 supports_function_calling: true - name: meta-llama/Meta-Llama-3.1-70B-Instruct max_input_tokens: 128000 - input_price: 0.52 - output_price: 0.75 + input_price: 0.35 + output_price: 0.4 supports_function_calling: true - name: meta-llama/Meta-Llama-3.1-8B-Instruct max_input_tokens: 128000 - input_price: 0.09 - output_price: 0.09 + input_price: 0.055 + output_price: 0.055 supports_function_calling: true - name: meta-llama/Meta-Llama-3-70B-Instruct max_input_tokens: 8192 - input_price: 0.59 - output_price: 0.79 + input_price: 0.35 + output_price: 0.4 - name: meta-llama/Meta-Llama-3-8B-Instruct max_input_tokens: 8192 - input_price: 0.08 - output_price: 0.08 + input_price: 0.055 + output_price: 0.055 - name: mistralai/Mistral-Nemo-Instruct-2407 max_input_tokens: 128000 input_price: 0.13 @@ -924,12 +953,12 @@ output_price: 0.27 - name: google/gemma-2-9b-it max_input_tokens: 8192 - input_price: 0.09 - output_price: 0.09 - - name: Qwen/Qwen2-72B-Instruct + input_price: 0.06 + output_price: 0.06 + - name: Qwen/Qwen2.5-72B-Instruct max_input_tokens: 32768 - input_price: 0.59 - output_price: 0.79 + input_price: 0.35 + output_price: 0.40 supports_function_calling: true - name: BAAI/bge-large-en-v1.5 type: embedding @@ -1174,18 +1203,27 @@ supports_function_calling: true - name: mistralai/mistral-large max_input_tokens: 128000 - input_price: 3 - output_price: 9 + input_price: 2 + output_price: 6 + supports_function_calling: true + - name: mistralai/mistral-small + input_price: 0.2 + output_price: 0.6 supports_function_calling: true - name: mistralai/mistral-nemo max_input_tokens: 128000 - input_price: 0.18 - output_price: 0.18 + input_price: 0.13 + output_price: 0.13 supports_function_calling: true - name: mistralai/codestral-mamba max_input_tokens: 256000 input_price: 0.25 output_price: 0.25 + - name: mistralai/pixtral-12b + max_input_tokens: 4096 + input_price: 0.1 + output_price: 0.1 + supports_vision: true - name: ai21/jamba-1-5-large max_input_tokens: 256000 input_price: 2 @@ -1196,22 +1234,22 @@ input_price: 0.2 output_price: 0.4 supports_function_calling: true - - name: cohere/command-r-plus + - name: cohere/command-r-plus-08-2024 max_input_tokens: 128000 input_price: 2.5 output_price: 10 supports_function_calling: true - - name: cohere/command-r-plus-08-2024 + - name: cohere/command-r-plus max_input_tokens: 128000 input_price: 2.5 output_price: 10 supports_function_calling: true - - name: cohere/command-r + - name: cohere/command-r-08-2024 max_input_tokens: 128000 input_price: 0.15 output_price: 0.6 supports_function_calling: true - - name: cohere/command-r-08-2024 + - name: cohere/command-r max_input_tokens: 128000 input_price: 0.15 output_price: 0.6 @@ -1221,11 +1259,6 @@ input_price: 0.14 output_price: 0.28 supports_function_calling: true - - name: deepseek/deepseek-coder - max_input_tokens: 32768 - input_price: 0.14 - output_price: 0.28 - supports_function_calling: true - name: perplexity/llama-3.1-sonar-huge-128k-online max_input_tokens: 127072 input_price: 5 @@ -1272,10 +1305,14 @@ max_input_tokens: 128000 input_price: 0.1 output_price: 0.1 - - name: qwen/qwen-2-72b-instruct + - name: qwen/qwen-2.5-72b-instruct + max_input_tokens: 131072 + input_price: 0.35 + output_price: 0.4 + - name: qwen/qwen-2-vl-72b-instruct max_input_tokens: 32768 - input_price: 0.9 - output_price: 0.9 + input_price: 0.4 + output_price: 0.4 # Links: # - https://octo.ai/docs/getting-started/inference-models @@ -1313,10 +1350,6 @@ # - https://docs.siliconflow.cn/reference/chat-completions-3 - platform: siliconflow models: - - name: Qwen/Qwen2-72B-Instruct - max_input_tokens: 32768 - input_price: 0 - output_price: 0 - name: meta-llama/Meta-Llama-3.1-405B-Instruct max_input_tokens: 32768 input_price: 2.94 @@ -1329,6 +1362,18 @@ max_input_tokens: 32768 input_price: 0 output_price: 0 + - name: Qwen/Qwen2.5-72B-Instruct + max_input_tokens: 32768 + input_price: 0.578 + output_price: 0.578 + - name: Qwen/Qwen2.5-7B-Instruct + max_input_tokens: 32768 + input_price: 0 + output_price: 0 + - name: Qwen/Qwen2.5-Coder-7B-Instruct + max_input_tokens: 32768 + input_price: 0 + output_price: 0 - name: google/gemma-2-27b-it max_input_tokens: 8192 input_price: 0.176 @@ -1341,10 +1386,6 @@ max_input_tokens: 32768 input_price: 0.186 output_price: 0.186 - - name: deepseek-ai/DeepSeek-Coder-V2-Instruct - max_input_tokens: 32768 - input_price: 0.186 - output_price: 0.186 - name: BAAI/bge-large-en-v1.5 type: embedding input_price: 0