chore(llma): update LLM costs (#39273)

This commit is contained in:
Radu Raicea
2025-10-07 14:56:32 -04:00
committed by GitHub
parent fa6e7115bc
commit 2385267c27
4 changed files with 141 additions and 135 deletions

View File

@@ -30,8 +30,8 @@
{
"model": "anubis-70b-v1.1",
"cost": {
"prompt_token": 4e-7,
"completion_token": 7e-7
"prompt_token": 6.5e-7,
"completion_token": 0.000001
}
},
{
@@ -151,9 +151,7 @@
"model": "claude-sonnet-4.5",
"cost": {
"prompt_token": 0.000003,
"completion_token": 0.000015,
"cache_read_token": 3e-7,
"cache_write_token": 0.00000375
"completion_token": 0.000015
}
},
{
@@ -258,8 +256,8 @@
{
"model": "deephermes-3-llama-3-8b-preview",
"cost": {
"prompt_token": 1e-8,
"completion_token": 5e-8
"prompt_token": 3e-8,
"completion_token": 1.1e-7
}
},
{
@@ -272,15 +270,15 @@
{
"model": "deephermes-3-mistral-24b-preview",
"cost": {
"prompt_token": 1.3e-7,
"completion_token": 5.1e-7
"prompt_token": 1.5e-7,
"completion_token": 5.9e-7
}
},
{
"model": "deepseek-chat",
"cost": {
"prompt_token": 2.4999988e-7,
"completion_token": 9.99999888e-7
"prompt_token": 3e-7,
"completion_token": 8.5e-7
}
},
{
@@ -335,8 +333,8 @@
{
"model": "deepseek-r1-0528-qwen3-8b",
"cost": {
"prompt_token": 1e-8,
"completion_token": 5e-8
"prompt_token": 3e-8,
"completion_token": 1.1e-7
}
},
{
@@ -398,8 +396,8 @@
{
"model": "deepseek-r1t-chimera",
"cost": {
"prompt_token": 2.5e-7,
"completion_token": 0.000001
"prompt_token": 3e-7,
"completion_token": 0.0000012
}
},
{
@@ -412,8 +410,8 @@
{
"model": "deepseek-r1t2-chimera",
"cost": {
"prompt_token": 2.5e-7,
"completion_token": 0.000001
"prompt_token": 3e-7,
"completion_token": 0.0000012
}
},
{
@@ -423,13 +421,6 @@
"completion_token": 0
}
},
{
"model": "deepseek-v3.1-base",
"cost": {
"prompt_token": 2.5e-7,
"completion_token": 0.000001
}
},
{
"model": "deepseek-v3.1-terminus",
"cost": {
@@ -461,8 +452,8 @@
{
"model": "devstral-small-2505",
"cost": {
"prompt_token": 4e-8,
"completion_token": 1.4e-7
"prompt_token": 5e-8,
"completion_token": 2.2e-7
}
},
{
@@ -482,8 +473,8 @@
{
"model": "dolphin3.0-mistral-24b",
"cost": {
"prompt_token": 3e-8,
"completion_token": 1.1e-7
"prompt_token": 4e-8,
"completion_token": 1.7e-7
}
},
{
@@ -493,20 +484,6 @@
"completion_token": 0
}
},
{
"model": "dolphin3.0-r1-mistral-24b",
"cost": {
"prompt_token": 1e-8,
"completion_token": 3e-8
}
},
{
"model": "dolphin3.0-r1-mistral-24b:free",
"cost": {
"prompt_token": 0,
"completion_token": 0
}
},
{
"model": "ernie-4.5-21b-a3b",
"cost": {
@@ -646,7 +623,7 @@
"model": "gemma-2-9b-it",
"cost": {
"prompt_token": 1e-8,
"completion_token": 2e-8
"completion_token": 3e-8
}
},
{
@@ -660,7 +637,7 @@
"model": "gemma-3-12b-it",
"cost": {
"prompt_token": 4e-8,
"completion_token": 1.4e-7
"completion_token": 1.3e-7
}
},
{
@@ -673,8 +650,8 @@
{
"model": "gemma-3-27b-it",
"cost": {
"prompt_token": 7e-8,
"completion_token": 2.6e-7
"prompt_token": 9e-8,
"completion_token": 1.6e-7
}
},
{
@@ -736,15 +713,15 @@
{
"model": "glm-4.5",
"cost": {
"prompt_token": 3.8e-7,
"completion_token": 0.0000016
"prompt_token": 3.5e-7,
"completion_token": 0.00000155
}
},
{
"model": "glm-4.5-air",
"cost": {
"prompt_token": 1.4e-7,
"completion_token": 8.6e-7
"prompt_token": 0,
"completion_token": 0
}
},
{
@@ -764,15 +741,15 @@
{
"model": "glm-4.6",
"cost": {
"prompt_token": 6e-7,
"completion_token": 0.000002
"prompt_token": 5e-7,
"completion_token": 0.00000175
}
},
{
"model": "glm-z1-32b",
"cost": {
"prompt_token": 4e-8,
"completion_token": 1.4e-7
"prompt_token": 5e-8,
"completion_token": 2.2e-7
}
},
{
@@ -984,6 +961,13 @@
"cache_read_token": 5e-9
}
},
{
"model": "gpt-5-pro",
"cost": {
"prompt_token": 0.000015,
"completion_token": 0.00012
}
},
{
"model": "gpt-oss-120b",
"cost": {
@@ -991,13 +975,6 @@
"completion_token": 4e-7
}
},
{
"model": "gpt-oss-120b:free",
"cost": {
"prompt_token": 0,
"completion_token": 0
}
},
{
"model": "gpt-oss-20b",
"cost": {
@@ -1060,13 +1037,6 @@
"cache_read_token": 5e-8
}
},
{
"model": "grok-4-fast:free",
"cost": {
"prompt_token": 0,
"completion_token": 0
}
},
{
"model": "grok-code-fast-1",
"cost": {
@@ -1099,8 +1069,8 @@
{
"model": "hermes-4-405b",
"cost": {
"prompt_token": 2.5e-7,
"completion_token": 0.000001
"prompt_token": 3e-7,
"completion_token": 0.0000012
}
},
{
@@ -1141,8 +1111,8 @@
{
"model": "internvl3-78b",
"cost": {
"prompt_token": 3e-8,
"completion_token": 1.3e-7
"prompt_token": 7e-8,
"completion_token": 2.6e-7
}
},
{
@@ -1183,8 +1153,8 @@
{
"model": "kimi-k2-0905",
"cost": {
"prompt_token": 4e-7,
"completion_token": 0.000002
"prompt_token": 3.9e-7,
"completion_token": 0.0000019
}
},
{
@@ -1194,20 +1164,6 @@
"completion_token": 0
}
},
{
"model": "kimi-vl-a3b-thinking",
"cost": {
"prompt_token": 2e-8,
"completion_token": 7e-8
}
},
{
"model": "kimi-vl-a3b-thinking:free",
"cost": {
"prompt_token": 0,
"completion_token": 0
}
},
{
"model": "l3-euryale-70b",
"cost": {
@@ -1264,18 +1220,11 @@
"completion_token": 6e-8
}
},
{
"model": "llama-3-lumimaid-70b",
"cost": {
"prompt_token": 0.000004,
"completion_token": 0.000006
}
},
{
"model": "llama-3.1-405b",
"cost": {
"prompt_token": 0.000002,
"completion_token": 0.000002
"prompt_token": 0.000004,
"completion_token": 0.000004
}
},
{
@@ -1439,6 +1388,13 @@
"completion_token": 7.5e-7
}
},
{
"model": "longcat-flash-chat:free",
"cost": {
"prompt_token": 0,
"completion_token": 0
}
},
{
"model": "maestro-reasoning",
"cost": {
@@ -1484,8 +1440,8 @@
{
"model": "mai-ds-r1",
"cost": {
"prompt_token": 2.5e-7,
"completion_token": 0.000001
"prompt_token": 3e-7,
"completion_token": 0.0000012
}
},
{
@@ -1631,8 +1587,8 @@
{
"model": "mistral-small-24b-instruct-2501",
"cost": {
"prompt_token": 4e-8,
"completion_token": 1.5e-7
"prompt_token": 5e-8,
"completion_token": 8e-8
}
},
{
@@ -1645,8 +1601,8 @@
{
"model": "mistral-small-3.1-24b-instruct",
"cost": {
"prompt_token": 4e-8,
"completion_token": 1.5e-7
"prompt_token": 5e-8,
"completion_token": 1e-7
}
},
{
@@ -1687,8 +1643,8 @@
{
"model": "mixtral-8x7b-instruct",
"cost": {
"prompt_token": 4e-7,
"completion_token": 4e-7
"prompt_token": 5.4e-7,
"completion_token": 5.4e-7
}
},
{
@@ -1927,7 +1883,7 @@
"model": "qwen-2.5-coder-32b-instruct",
"cost": {
"prompt_token": 4e-8,
"completion_token": 1.4e-7
"completion_token": 1.6e-7
}
},
{
@@ -1996,11 +1952,18 @@
"completion_token": 6.3e-7
}
},
{
"model": "qwen2.5-coder-7b-instruct",
"cost": {
"prompt_token": 3e-8,
"completion_token": 9e-8
}
},
{
"model": "qwen2.5-vl-32b-instruct",
"cost": {
"prompt_token": 4e-8,
"completion_token": 1.4e-7
"prompt_token": 5e-8,
"completion_token": 2.2e-7
}
},
{
@@ -2013,8 +1976,8 @@
{
"model": "qwen2.5-vl-72b-instruct",
"cost": {
"prompt_token": 7e-8,
"completion_token": 2.8e-7
"prompt_token": 8e-8,
"completion_token": 3.3e-7
}
},
{
@@ -2027,8 +1990,8 @@
{
"model": "qwen3-14b",
"cost": {
"prompt_token": 4e-8,
"completion_token": 1.4e-7
"prompt_token": 5e-8,
"completion_token": 2.2e-7
}
},
{
@@ -2076,8 +2039,8 @@
{
"model": "qwen3-30b-a3b-instruct-2507",
"cost": {
"prompt_token": 7e-8,
"completion_token": 2.8e-7
"prompt_token": 8e-8,
"completion_token": 3.3e-7
}
},
{
@@ -2097,8 +2060,8 @@
{
"model": "qwen3-32b",
"cost": {
"prompt_token": 3e-8,
"completion_token": 1.3e-7
"prompt_token": 5e-8,
"completion_token": 2e-7
}
},
{
@@ -2177,15 +2140,15 @@
{
"model": "qwen3-next-80b-a3b-thinking",
"cost": {
"prompt_token": 1e-7,
"completion_token": 8e-7
"prompt_token": 1.4e-7,
"completion_token": 0.0000012
}
},
{
"model": "qwen3-vl-235b-a22b-instruct",
"cost": {
"prompt_token": 3e-7,
"completion_token": 0.0000015
"completion_token": 0.0000012
}
},
{
@@ -2205,8 +2168,8 @@
{
"model": "qwq-32b-arliai-rpr-v1",
"cost": {
"prompt_token": 2e-8,
"completion_token": 7e-8
"prompt_token": 3e-8,
"completion_token": 1.1e-7
}
},
{
@@ -2251,18 +2214,11 @@
"completion_token": 0.0000034
}
},
{
"model": "seed-oss-36b-instruct",
"cost": {
"prompt_token": 1.6e-7,
"completion_token": 6.5e-7
}
},
{
"model": "shisa-v2-llama3.3-70b",
"cost": {
"prompt_token": 4e-8,
"completion_token": 1.4e-7
"prompt_token": 5e-8,
"completion_token": 2.2e-7
}
},
{
@@ -2275,8 +2231,8 @@
{
"model": "skyfall-36b-v2",
"cost": {
"prompt_token": 4e-8,
"completion_token": 1.6e-7
"prompt_token": 8e-8,
"completion_token": 3.3e-7
}
},
{
@@ -2342,6 +2298,13 @@
"completion_token": 4e-7
}
},
{
"model": "tongyi-deepresearch-30b-a3b:free",
"cost": {
"prompt_token": 0,
"completion_token": 0
}
},
{
"model": "ui-tars-1.5-7b",
"cost": {

View File

@@ -8250,7 +8250,8 @@
"provider": "openrouter",
"cost": {
"prompt_token": 0.000003,
"completion_token": 0.000015
"completion_token": 0.000015,
"cache_read_token": 7.5e-7
}
},
{
@@ -8266,7 +8267,8 @@
"provider": "openrouter",
"cost": {
"prompt_token": 3e-7,
"completion_token": 5e-7
"completion_token": 5e-7,
"cache_read_token": 8e-8
}
},
{
@@ -8282,7 +8284,17 @@
"provider": "openrouter",
"cost": {
"prompt_token": 0.000003,
"completion_token": 0.000015
"completion_token": 0.000015,
"cache_read_token": 7.5e-7
}
},
{
"model": "x-ai/grok-4-fast",
"provider": "openrouter",
"cost": {
"prompt_token": 2e-7,
"completion_token": 5e-7,
"cache_read_token": 5e-8
}
},
{

View File

@@ -0,0 +1,14 @@
import { manualCosts } from './manual-providers'
describe('manualCosts', () => {
describe('openrouter/auto model', () => {
it('is defined with zero costs', () => {
const openrouterAuto = manualCosts.find((model) => model.model === 'openrouter/auto')
expect(openrouterAuto).toBeDefined()
expect(openrouterAuto?.provider).toBe('openrouter')
expect(openrouterAuto?.cost.prompt_token).toBe(0)
expect(openrouterAuto?.cost.completion_token).toBe(0)
})
})
})

View File

@@ -66,4 +66,21 @@ export const manualCosts: ModelRow[] = [
completion_token: 0.0000009,
},
},
{
model: 'mistral-medium-3',
provider: 'mistral',
cost: {
prompt_token: 4e-7,
completion_token: 0.000002,
},
},
// We can't know the model, so we set the cost to 0
{
model: 'openrouter/auto',
provider: 'openrouter',
cost: {
prompt_token: 0,
completion_token: 0,
},
},
]