From b308ff004d0aeaf84732cef5a3f9283dd725ae58 Mon Sep 17 00:00:00 2001 From: Enrique Cardoza Date: Mon, 5 Aug 2024 15:07:50 -0500 Subject: [PATCH] Update TogetherAI costs (#2380) * update groq costs * update togetherai costs --- costs/__tests__/ensureOnlyOne.ts | 7 ++ costs/src/providers/togetherai/chat/llama.ts | 70 +++++++++++++++++++ .../cost/providers/togetherai/chat/llama.ts | 70 +++++++++++++++++++ .../cost/providers/togetherai/chat/llama.ts | 70 +++++++++++++++++++ .../cost/providers/togetherai/chat/llama.ts | 70 +++++++++++++++++++ 5 files changed, 287 insertions(+) diff --git a/costs/__tests__/ensureOnlyOne.ts b/costs/__tests__/ensureOnlyOne.ts index 2a1d6e5841..fe38870b38 100644 --- a/costs/__tests__/ensureOnlyOne.ts +++ b/costs/__tests__/ensureOnlyOne.ts @@ -375,6 +375,13 @@ WHEN (request_response_log.model = 'meta-llama/Llama-3-8b-chat-hf') THEN 200 * r WHEN (request_response_log.model = 'NousResearch/Nous-Hermes-llama-2-7b') THEN 200 * request_response_log.prompt_tokens + 200 * request_response_log.completion_tokens WHEN (request_response_log.model = 'NousResearch/Nous-Hermes-Llama2-13b') THEN 225 * request_response_log.prompt_tokens + 225 * request_response_log.completion_tokens WHEN (request_response_log.model = 'togethercomputer/Llama-2-7B-32K-Instruct') THEN 200 * request_response_log.prompt_tokens + 200 * request_response_log.completion_tokens +WHEN (request_response_log.model = 'meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo') THEN 880 * request_response_log.prompt_tokens + 880 * request_response_log.completion_tokens +WHEN (request_response_log.model = 'meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo') THEN 180 * request_response_log.prompt_tokens + 180 * request_response_log.completion_tokens +WHEN (request_response_log.model = 'meta-llama/Meta-Llama-3.1-405B-Instruct-Turbo') THEN 5000 * request_response_log.prompt_tokens + 5000 * request_response_log.completion_tokens +WHEN (request_response_log.model = 'meta-llama/Meta-Llama-3-70B-Instruct-Turbo') THEN 880 * request_response_log.prompt_tokens + 880 * request_response_log.completion_tokens +WHEN (request_response_log.model = 'meta-llama/Meta-Llama-3-8B-Instruct-Turbo') THEN 180 * request_response_log.prompt_tokens + 180 * request_response_log.completion_tokens +WHEN (request_response_log.model = 'meta-llama/Meta-Llama-3-70B-Instruct-Lite') THEN 540 * request_response_log.prompt_tokens + 540 * request_response_log.completion_tokens +WHEN (request_response_log.model = 'meta-llama/Meta-Llama-3-8B-Instruct-Lite') THEN 100 * request_response_log.prompt_tokens + 100 * request_response_log.completion_tokens WHEN (request_response_log.model = 'zero-one-ai/Yi-34B') THEN 800 * request_response_log.prompt_tokens + 800 * request_response_log.completion_tokens WHEN (request_response_log.model = 'zero-one-ai/Yi-6B') THEN 200 * request_response_log.prompt_tokens + 200 * request_response_log.completion_tokens WHEN (request_response_log.model = 'google/gemma-2b') THEN 100 * request_response_log.prompt_tokens + 100 * request_response_log.completion_tokens diff --git a/costs/src/providers/togetherai/chat/llama.ts b/costs/src/providers/togetherai/chat/llama.ts index b8a0779197..32db21c26d 100644 --- a/costs/src/providers/togetherai/chat/llama.ts +++ b/costs/src/providers/togetherai/chat/llama.ts @@ -137,4 +137,74 @@ export const costs: ModelRow[] = [ completion_token: 0.0000002, }, }, + { + model: { + operator: "equals", + value: "meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo", + }, + cost: { + prompt_token: 0.00000088, + completion_token: 0.00000088, + }, + }, + { + model: { + operator: "equals", + value: "meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo", + }, + cost: { + prompt_token: 0.00000018, + completion_token: 0.00000018, + }, + }, + { + model: { + operator: "equals", + value: "meta-llama/Meta-Llama-3.1-405B-Instruct-Turbo", + }, + cost: { + prompt_token: 0.000005, + completion_token: 0.000005, + }, + }, + { + model: { + operator: "equals", + value: "meta-llama/Meta-Llama-3-70B-Instruct-Turbo", + }, + cost: { + prompt_token: 0.00000088, + completion_token: 0.00000088, + }, + }, + { + model: { + operator: "equals", + value: "meta-llama/Meta-Llama-3-8B-Instruct-Turbo", + }, + cost: { + prompt_token: 0.00000018, + completion_token: 0.00000018, + }, + }, + { + model: { + operator: "equals", + value: "meta-llama/Meta-Llama-3-70B-Instruct-Lite", + }, + cost: { + prompt_token: 0.00000054, + completion_token: 0.00000054, + }, + }, + { + model: { + operator: "equals", + value: "meta-llama/Meta-Llama-3-8B-Instruct-Lite", + }, + cost: { + prompt_token: 0.0000001, + completion_token: 0.0000001, + }, + }, ]; diff --git a/valhalla/jawn/src/packages/cost/providers/togetherai/chat/llama.ts b/valhalla/jawn/src/packages/cost/providers/togetherai/chat/llama.ts index b8a0779197..32db21c26d 100644 --- a/valhalla/jawn/src/packages/cost/providers/togetherai/chat/llama.ts +++ b/valhalla/jawn/src/packages/cost/providers/togetherai/chat/llama.ts @@ -137,4 +137,74 @@ export const costs: ModelRow[] = [ completion_token: 0.0000002, }, }, + { + model: { + operator: "equals", + value: "meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo", + }, + cost: { + prompt_token: 0.00000088, + completion_token: 0.00000088, + }, + }, + { + model: { + operator: "equals", + value: "meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo", + }, + cost: { + prompt_token: 0.00000018, + completion_token: 0.00000018, + }, + }, + { + model: { + operator: "equals", + value: "meta-llama/Meta-Llama-3.1-405B-Instruct-Turbo", + }, + cost: { + prompt_token: 0.000005, + completion_token: 0.000005, + }, + }, + { + model: { + operator: "equals", + value: "meta-llama/Meta-Llama-3-70B-Instruct-Turbo", + }, + cost: { + prompt_token: 0.00000088, + completion_token: 0.00000088, + }, + }, + { + model: { + operator: "equals", + value: "meta-llama/Meta-Llama-3-8B-Instruct-Turbo", + }, + cost: { + prompt_token: 0.00000018, + completion_token: 0.00000018, + }, + }, + { + model: { + operator: "equals", + value: "meta-llama/Meta-Llama-3-70B-Instruct-Lite", + }, + cost: { + prompt_token: 0.00000054, + completion_token: 0.00000054, + }, + }, + { + model: { + operator: "equals", + value: "meta-llama/Meta-Llama-3-8B-Instruct-Lite", + }, + cost: { + prompt_token: 0.0000001, + completion_token: 0.0000001, + }, + }, ]; diff --git a/web/packages/cost/providers/togetherai/chat/llama.ts b/web/packages/cost/providers/togetherai/chat/llama.ts index b8a0779197..32db21c26d 100644 --- a/web/packages/cost/providers/togetherai/chat/llama.ts +++ b/web/packages/cost/providers/togetherai/chat/llama.ts @@ -137,4 +137,74 @@ export const costs: ModelRow[] = [ completion_token: 0.0000002, }, }, + { + model: { + operator: "equals", + value: "meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo", + }, + cost: { + prompt_token: 0.00000088, + completion_token: 0.00000088, + }, + }, + { + model: { + operator: "equals", + value: "meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo", + }, + cost: { + prompt_token: 0.00000018, + completion_token: 0.00000018, + }, + }, + { + model: { + operator: "equals", + value: "meta-llama/Meta-Llama-3.1-405B-Instruct-Turbo", + }, + cost: { + prompt_token: 0.000005, + completion_token: 0.000005, + }, + }, + { + model: { + operator: "equals", + value: "meta-llama/Meta-Llama-3-70B-Instruct-Turbo", + }, + cost: { + prompt_token: 0.00000088, + completion_token: 0.00000088, + }, + }, + { + model: { + operator: "equals", + value: "meta-llama/Meta-Llama-3-8B-Instruct-Turbo", + }, + cost: { + prompt_token: 0.00000018, + completion_token: 0.00000018, + }, + }, + { + model: { + operator: "equals", + value: "meta-llama/Meta-Llama-3-70B-Instruct-Lite", + }, + cost: { + prompt_token: 0.00000054, + completion_token: 0.00000054, + }, + }, + { + model: { + operator: "equals", + value: "meta-llama/Meta-Llama-3-8B-Instruct-Lite", + }, + cost: { + prompt_token: 0.0000001, + completion_token: 0.0000001, + }, + }, ]; diff --git a/worker/src/packages/cost/providers/togetherai/chat/llama.ts b/worker/src/packages/cost/providers/togetherai/chat/llama.ts index b8a0779197..32db21c26d 100644 --- a/worker/src/packages/cost/providers/togetherai/chat/llama.ts +++ b/worker/src/packages/cost/providers/togetherai/chat/llama.ts @@ -137,4 +137,74 @@ export const costs: ModelRow[] = [ completion_token: 0.0000002, }, }, + { + model: { + operator: "equals", + value: "meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo", + }, + cost: { + prompt_token: 0.00000088, + completion_token: 0.00000088, + }, + }, + { + model: { + operator: "equals", + value: "meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo", + }, + cost: { + prompt_token: 0.00000018, + completion_token: 0.00000018, + }, + }, + { + model: { + operator: "equals", + value: "meta-llama/Meta-Llama-3.1-405B-Instruct-Turbo", + }, + cost: { + prompt_token: 0.000005, + completion_token: 0.000005, + }, + }, + { + model: { + operator: "equals", + value: "meta-llama/Meta-Llama-3-70B-Instruct-Turbo", + }, + cost: { + prompt_token: 0.00000088, + completion_token: 0.00000088, + }, + }, + { + model: { + operator: "equals", + value: "meta-llama/Meta-Llama-3-8B-Instruct-Turbo", + }, + cost: { + prompt_token: 0.00000018, + completion_token: 0.00000018, + }, + }, + { + model: { + operator: "equals", + value: "meta-llama/Meta-Llama-3-70B-Instruct-Lite", + }, + cost: { + prompt_token: 0.00000054, + completion_token: 0.00000054, + }, + }, + { + model: { + operator: "equals", + value: "meta-llama/Meta-Llama-3-8B-Instruct-Lite", + }, + cost: { + prompt_token: 0.0000001, + completion_token: 0.0000001, + }, + }, ];