diff options
| -rw-r--r-- | models.yaml | 346 | ||||
| -rw-r--r-- | src/client/claude.rs | 3 |
2 files changed, 157 insertions, 192 deletions
diff --git a/models.yaml b/models.yaml index ac20e1e..2ac3da9 100644 --- a/models.yaml +++ b/models.yaml @@ -1,6 +1,5 @@ # notes: # - do not submit pull requests to add new models; this list will be updated in batches with new releases. -# - do not add any open-source LLMs except for the following: Mixtral, LLama-3, Gemma, Qwen, Phi-3, DeepSeek, Command-R, dbrx, Yi. - platform: openai # docs: @@ -10,18 +9,6 @@ # notes # - get max_output_tokens info from api error models: - - name: gpt-3.5-turbo - max_input_tokens: 16385 - max_output_tokens: 4096 - input_price: 0.5 - output_price: 1.5 - supports_function_calling: true - - name: gpt-3.5-turbo-1106 - max_input_tokens: 16385 - max_output_tokens: 4096 - input_price: 1 - output_price: 2 - supports_function_calling: true - name: gpt-4o max_input_tokens: 128000 max_output_tokens: 4096 @@ -48,23 +35,23 @@ input_price: 10 output_price: 30 supports_function_calling: true - - name: gpt-4-vision-preview - max_input_tokens: 128000 - max_output_tokens: 4096 - require_max_tokens: true - input_price: 10 - output_price: 30 - supports_vision: true - name: gpt-4 max_input_tokens: 8192 max_output_tokens: 4096 input_price: 30 output_price: 60 - - name: gpt-4-32k - max_input_tokens: 32768 + - name: gpt-3.5-turbo + max_input_tokens: 16385 + max_output_tokens: 4096 + input_price: 0.5 + output_price: 1.5 + supports_function_calling: true + - name: gpt-3.5-turbo-1106 + max_input_tokens: 16385 max_output_tokens: 4096 - input_price: 60 - output_price: 120 + input_price: 1 + output_price: 2 + supports_function_calling: true - name: text-embedding-3-large mode: embedding max_input_tokens: 8191 @@ -84,32 +71,26 @@ # notes: # - get max_output_tokens info from list models api models: - - name: gemini-1.0-pro-latest - max_input_tokens: 30720 - max_output_tokens: 2048 - input_price: 0.5 - output_price: 1.5 - supports_function_calling: true - - name: gemini-1.0-pro-vision-latest - max_input_tokens: 12288 - max_output_tokens: 4096 - input_price: 0.5 - output_price: 1.5 - supports_vision: true - - name: gemini-1.5-flash-latest + - name: gemini-1.5-pro-latest max_input_tokens: 1048576 max_output_tokens: 8192 - input_price: 0.35 - output_price: 0.53 + input_price: 3.5 + output_price: 10.5 supports_vision: true supports_function_calling: true - - name: gemini-1.5-pro-latest + - name: gemini-1.5-flash-latest max_input_tokens: 1048576 max_output_tokens: 8192 - input_price: 3.5 - output_price: 10.5 + input_price: 0.35 + output_price: 1.05 supports_vision: true supports_function_calling: true + - name: gemini-1.0-pro-latest + max_input_tokens: 30720 + max_output_tokens: 2048 + input_price: 0.5 + output_price: 1.5 + supports_function_calling: true - name: text-embedding-004 mode: embedding max_input_tokens: 2048 @@ -167,15 +148,21 @@ max_input_tokens: 64000 input_price: 2 output_price: 6 + supports_function_calling: true - name: mistral-small-latest max_input_tokens: 32000 - input_price: 2 - output_price: 6 + input_price: 1 + output_price: 3 + supports_function_calling: true - name: mistral-large-latest max_input_tokens: 32000 - input_price: 8 - output_price: 24 + input_price: 4 + output_price: 12 supports_function_calling: true + - name: codestral-latest + max_input_tokens: 32000 + input_price: 1 + output_price: 3 - name: mistral-embed mode: embedding max_input_tokens: 8092 @@ -230,7 +217,6 @@ max_output_tokens: 32768 input_price: 0.6 output_price: 0.6 - - name: llama-3-8b-instruct max_input_tokens: 8192 max_output_tokens: 8192 @@ -271,11 +257,6 @@ max_output_tokens: 32768 input_price: 0.27 output_price: 0.27 - - name: gemma-7b-it - max_input_tokens: 8192 - max_output_tokens: 8192 - input_price: 0.10 - output_price: 0.10 - platform: vertexai # docs: @@ -285,30 +266,26 @@ # notes: # - get max_output_tokens info from models doc models: - - name: gemini-1.0-pro-002 - max_input_tokens: 24568 + - name: gemini-1.5-pro-001 + max_input_tokens: 1000000 max_output_tokens: 8192 - input_price: 0.125 - output_price: 0.375 - supports_function_calling: true - - name: gemini-1.0-pro-vision-001 - max_input_tokens: 14336 - max_output_tokens: 2048 - input_price: 0.125 - output_price: 0.375 + input_price: 1.25 + output_price: 3.75 supports_vision: true - - name: gemini-1.5-flash-preview-0514 + supports_function_calling: true + - name: gemini-1.5-flash-001 max_input_tokens: 1000000 max_output_tokens: 8192 input_price: 0.125 output_price: 0.375 supports_vision: true - - name: gemini-1.5-pro-preview-0514 - max_input_tokens: 1000000 + supports_function_calling: true + - name: gemini-1.0-pro-002 + max_input_tokens: 24568 max_output_tokens: 8192 - input_price: 1.25 - output_price: 3.75 - supports_vision: true + input_price: 0.125 + output_price: 0.375 + supports_function_calling: true - name: text-embedding-004 mode: embedding max_input_tokens: 3072 @@ -334,6 +311,7 @@ input_price: 15 output_price: 75 supports_vision: true + supports_function_calling: true - name: claude-3-sonnet@20240229 max_input_tokens: 200000 max_output_tokens: 4096 @@ -341,6 +319,7 @@ input_price: 3 output_price: 15 supports_vision: true + supports_function_calling: true - name: claude-3-haiku@20240307 max_input_tokens: 200000 max_output_tokens: 4096 @@ -348,6 +327,7 @@ input_price: 0.25 output_price: 1.25 supports_vision: true + supports_function_calling: true - platform: bedrock # docs: @@ -364,6 +344,7 @@ input_price: 15 output_price: 75 supports_vision: true + supports_function_calling: true - name: anthropic.claude-3-sonnet-20240229-v1:0 max_input_tokens: 200000 max_output_tokens: 4096 @@ -371,6 +352,7 @@ input_price: 3 output_price: 15 supports_vision: true + supports_function_calling: true - name: anthropic.claude-3-haiku-20240307-v1:0 max_input_tokens: 200000 max_output_tokens: 4096 @@ -378,6 +360,7 @@ input_price: 0.25 output_price: 1.25 supports_vision: true + supports_function_calling: true - name: meta.llama3-8b-instruct-v1:0 max_input_tokens: 8192 max_output_tokens: 2048 @@ -424,22 +407,14 @@ max_input_tokens: 4096 max_output_tokens: 4096 require_max_tokens: true - - name: '@cf/google/gemma-7b-it-lora' - max_input_tokens: 4096 - max_output_tokens: 4096 - require_max_tokens: true - name: '@cf/qwen/qwen1.5-14b-chat-awq' max_input_tokens: 4096 max_output_tokens: 4096 require_max_tokens: true - - name: '@hf/thebloke/deepseek-coder-6.7b-instruct-awq' - max_input_tokens: 4096 - max_output_tokens: 4096 - require_max_tokens: true - platform: replicate # docs: - # - https://replicate.com/docs + # - https://replicate.com/explore # - https://replicate.com/pricing # - https://replicate.com/docs/reference/http # notes: @@ -477,13 +452,13 @@ # notes: # - get max_output_tokens info from models doc models: - - name: ernie-4.0-8k-preview + - name: ernie-4.0-8k-0613 max_input_tokens: 5120 max_output_tokens: 2048 require_max_tokens: true input_price: 16.8 output_price: 16.8 - - name: ernie-3.5-8k-preview + - name: ernie-3.5-8k-0613 max_input_tokens: 5120 max_output_tokens: 2048 require_max_tokens: true @@ -501,12 +476,6 @@ require_max_tokens: true input_price: 0 output_price: 0 - - name: ernie-tiny-8k - max_input_tokens: 7168 - max_output_tokens: 2048 - require_max_tokens: true - input_price: 0 - output_price: 0 - platform: qianwen # docs: @@ -584,7 +553,7 @@ input_price: 0.14 output_price: 0.28 - name: deepseek-coder - max_input_tokens: 16384 + max_input_tokens: 32768 input_price: 0.14 output_price: 0.28 @@ -593,23 +562,35 @@ # - https://open.bigmodel.cn/dev/howuse/model # - https://open.bigmodel.cn/pricing models: - - name: glm-4 + - name: glm-4-0520 max_input_tokens: 128000 input_price: 14 output_price: 14 + supports_function_calling: true + - name: glm-4-airx + max_input_tokens: 8092 + input_price: 1.4 + output_price: 1.4 + supports_function_calling: true + - name: glm-4-air + max_input_tokens: 128000 + input_price: 0.14 + output_price: 0.14 + supports_function_calling: true + - name: glm-4-flash + max_input_tokens: 128000 + input_price: 0.014 + output_price: 0.014 + supports_function_calling: true - name: glm-4v max_input_tokens: 2048 - input_price: 14 - output_price: 14 + input_price: 7 + output_price: 7 supports_vision: true - - name: glm-3-turbo - max_input_tokens: 128000 - input_price: 0.7 - output_price: 0.7 - platform: anyscale # docs: - # - https://docs.endpoints.anyscale.com/text-generation/query-a-model/ + # - https://docs.anyscale.com/endpoints/text-generation/supported-models/meta-llama-Meta-Llama-3-8B-Instruct # - https://docs.endpoints.anyscale.com/pricing models: - name: meta-llama/Meta-Llama-3-8B-Instruct @@ -636,10 +617,6 @@ max_input_tokens: 65536 input_price: 0.90 output_price: 0.90 - - name: google/gemma-7b-it - max_input_tokens: 8192 - input_price: 0.15 - output_price: 0.15 - platform: deepinfra # docs: @@ -654,7 +631,7 @@ max_input_tokens: 8192 input_price: 0.59 output_price: 0.79 - - name: mistralai/Mistral-7B-Instruct-v0.2 + - name: mistralai/Mistral-7B-Instruct-v0.3 max_input_tokens: 32768 input_price: 0.07 output_price: 0.07 @@ -666,24 +643,30 @@ max_input_tokens: 65536 input_price: 0.65 output_price: 0.65 - - name: google/gemma-1.1-7b-it - max_input_tokens: 8192 - input_price: 0.07 - output_price: 0.07 - - name: databricks/dbrx-instruct + - name: Qwen/Qwen2-72B-Instruct max_input_tokens: 32768 - input_price: 0.6 - output_price: 0.6 - - name: 01-ai/Yi-34B-Chat + input_price: 0.59 + output_price: 0.79 + - name: microsoft/Phi-3-medium-4k-instruct max_input_tokens: 4096 - input_price: 0.6 - output_price: 0.6 + input_price: 0.14 + output_price: 0.14 - platform: fireworks # docs: # - https://fireworks.ai/models # - https://fireworks.ai/pricing models: + - name: accounts/fireworks/models/firellava-13b + max_input_tokens: 4096 + input_price: 0.2 + output_price: 0.2 + supports_vision: true + - name: accounts/fireworks/models/firefunction-v1 + max_input_tokens: 32768 + input_price: 0.2 + output_price: 0.2 + supports_function_calling: true - name: accounts/fireworks/models/llama-v3-8b-instruct max_input_tokens: 8192 input_price: 0.2 @@ -692,7 +675,7 @@ max_input_tokens: 8192 input_price: 0.9 output_price: 0.9 - - name: accounts/fireworks/models/mistral-7b-instruct-v0p2 + - name: accounts/fireworks/models/mistral-7b-instruct-v3 max_input_tokens: 32768 input_price: 0.2 output_price: 0.2 @@ -704,18 +687,19 @@ max_input_tokens: 65536 input_price: 0.9 output_price: 0.9 - - name: accounts/fireworks/models/qwen-72b-chat - max_input_tokens: 4096 + - name: accounts/fireworks/models/qwen2-72b-instruct + max_input_tokens: 32768 input_price: 0.9 output_price: 0.9 - - name: accounts/fireworks/models/gemma-7b-it - max_input_tokens: 8192 + - name: accounts/fireworks/models/phi-3-mini-128k-instruct + max_input_tokens: 131072 input_price: 0.2 output_price: 0.2 - - name: accounts/fireworks/models/dbrx-instruct - max_input_tokens: 32768 - input_price: 1.6 - output_price: 1.6 + - name: accounts/fireworks/models/phi-3-vision-128k-instruct + max_input_tokens: 131072 + input_price: 0.2 + output_price: 0.2 + supports_vision: true - platform: openrouter # docs: @@ -723,45 +707,36 @@ models: - name: meta-llama/llama-3-8b-instruct max_input_tokens: 8192 - input_price: 0.1 - output_price: 0.1 + input_price: 0.07 + output_price: 0.07 - name: meta-llama/llama-3-8b-instruct:nitro max_input_tokens: 8192 input_price: 0.2 output_price: 0.2 - name: meta-llama/llama-3-8b-instruct:extended max_input_tokens: 16384 - input_price: 0.275 - output_price: 0.283 + input_price: 0.2 + output_price: 1.125 - name: meta-llama/llama-3-70b-instruct max_input_tokens: 8192 - input_price: 0.81 - output_price: 0.81 + input_price: 0.59 + output_price: 0.79 - name: meta-llama/llama-3-70b-instruct:nitro max_input_tokens: 8192 input_price: 0.9 output_price: 0.9 - - name: mistralai/mistral-7b-instruct:free + - name: mistralai/mistral-7b-instruct-v0.3 max_input_tokens: 32768 - input_price: 0.0 - output_price: 0.0 - - name: codellama/codellama-70b-instruct - max_input_tokens: 2048 - input_price: 0.81 - output_price: 0.81 - - name: google/gemma-7b-it:free - max_input_tokens: 8192 - input_price: 0.0 - output_price: 0.0 - - name: 01-ai/yi-34b-chat - max_input_tokens: 4096 - input_price: 0.72 - output_price: 0.72 - - name: openai/gpt-3.5-turbo - max_input_tokens: 16385 - input_price: 0.5 - output_price: 1.5 - supports_function_calling: true + input_price: 0.07 + output_price: 0.07 + - name: microsoft/phi-3-mini-128k-instruct + max_input_tokens: 128000 + input_price: 0.1 + output_price: 0.1 + - name: qwen/qwen-2-72b-instruct + max_input_tokens: 32768 + input_price: 0.9 + output_price: 0.9 - name: openai/gpt-4o max_input_tokens: 128000 input_price: 5 @@ -779,34 +754,32 @@ input_price: 10 output_price: 30 supports_function_calling: true - - name: openai/gpt-4-vision-preview - max_input_tokens: 128000 - max_output_tokens: 4096 - input_price: 10 - output_price: 30 - supports_vision: true - name: openai/gpt-4 max_input_tokens: 8192 input_price: 30 output_price: 60 - - name: openai/gpt-4-32k - max_input_tokens: 32768 - input_price: 60 - output_price: 120 - - name: google/gemini-pro - max_input_tokens: 91728 - input_price: 0.125 - output_price: 0.375 - - name: google/gemini-pro-vision - max_input_tokens: 45875 - input_price: 0.125 - output_price: 0.375 - supports_vision: true + - name: openai/gpt-3.5-turbo + max_input_tokens: 16385 + input_price: 0.5 + output_price: 1.5 + supports_function_calling: true - name: google/gemini-pro-1.5 max_input_tokens: 2800000 input_price: 2.5 output_price: 7.5 supports_vision: true + supports_function_calling: true + - name: google/gemini-flash-1.5 + max_input_tokens: 2800000 + input_price: 0.25 + output_price: 0.75 + supports_vision: true + supports_function_calling: true + - name: google/gemini-pro + max_input_tokens: 91728 + input_price: 0.125 + output_price: 0.375 + supports_function_calling: true - name: anthropic/claude-3-opus max_input_tokens: 200000 max_output_tokens: 4096 @@ -814,6 +787,7 @@ input_price: 15 output_price: 75 supports_vision: true + supports_function_calling: true - name: anthropic/claude-3-sonnet max_input_tokens: 200000 max_output_tokens: 4096 @@ -821,6 +795,7 @@ input_price: 3 output_price: 15 supports_vision: true + supports_function_calling: true - name: anthropic/claude-3-haiku max_input_tokens: 200000 max_output_tokens: 4096 @@ -828,6 +803,7 @@ input_price: 0.25 output_price: 1.25 supports_vision: true + supports_function_calling: true - name: mistralai/mixtral-8x7b-instruct max_input_tokens: 32768 input_price: 0.24 @@ -836,26 +812,36 @@ max_input_tokens: 65536 input_price: 0.65 output_price: 0.65 + supports_function_calling: true - name: mistralai/mistral-small max_input_tokens: 32000 input_price: 2 output_price: 6 + supports_function_calling: true - name: mistralai/mistral-large max_input_tokens: 32000 input_price: 8 output_price: 24 - - name: databricks/dbrx-instruct - max_input_tokens: 32768 - input_price: 0.6 - output_price: 0.6 + supports_function_calling: true - name: cohere/command-r max_input_tokens: 128000 input_price: 0.5 output_price: 1.5 + supports_function_calling: true - name: cohere/command-r-plus max_input_tokens: 128000 input_price: 3 output_price: 15 + supports_function_calling: true + - name: deepseek/deepseek-chat + max_input_tokens: 32768 + input_price: 0.14 + output_price: 0.28 + - name: deepseek/deepseek-coder + max_input_tokens: 32768 + input_price: 0.14 + output_price: 0.28 + - platform: octoai # docs: @@ -896,7 +882,7 @@ max_input_tokens: 8000 input_price: 0.9 output_price: 0.9 - - name: mistralai/Mistral-7B-Instruct-v0.2 + - name: mistralai/Mistral-7B-Instruct-v0.3 max_input_tokens: 32768 input_price: 0.2 output_price: 0.2 @@ -908,27 +894,7 @@ max_input_tokens: 65536 input_price: 1.2 output_price: 1.2 - - name: google/gemma-7b-it - max_input_tokens: 8192 - input_price: 0.2 - output_price: 0.2 - - name: Qwen/Qwen1.5-72B-Chat - max_input_tokens: 32768 - input_price: 0.9 - output_price: 0.9 - - name: databricks/dbrx-instruct + - name: Qwen/Qwen2-72B-Instruct max_input_tokens: 32768 - input_price: 1.2 - output_price: 1.2 - - name: zero-one-ai/Yi-34B-Chat - max_input_tokens: 4096 - input_price: 0.8 - output_price: 0.8 - - name: deepseek-ai/deepseek-llm-67b-chat - max_input_tokens: 4096 input_price: 0.9 - output_price: 0.9 - - name: deepseek-ai/deepseek-coder-33b-instruct - max_input_tokens: 16384 - input_price: 0.8 - output_price: 0.8 + output_price: 0.9
\ No newline at end of file diff --git a/src/client/claude.rs b/src/client/claude.rs index a16722b..f95fafc 100644 --- a/src/client/claude.rs +++ b/src/client/claude.rs @@ -39,8 +39,7 @@ impl ClaudeClient { let mut builder = client.post(url).json(&body); builder = builder - .header("anthropic-version", "2023-06-01") - .header("anthropic-beta", "tools-2024-05-16"); + .header("anthropic-version", "2023-06-01"); if let Some(api_key) = api_key { builder = builder.header("x-api-key", api_key) } |