mirror of
https://github.com/BerriAI/litellm.git
synced 2026-10-03 02:22:24 +00:00
Fix us-gov-{east,west}-1 Bedrock Sonnet 4.5 pricing (Fixes #27120)
AWS GovCloud Bedrock pricing carries a +20% premium over the global Anthropic rates, not the +10% premium that applies to commercial US regions. Litellm's five us-gov Claude Sonnet 4.5 entries mirror the commercial-US prices (3.3e-06 input, 1.65e-05 output, 4.125e-06 cache write), undercharging customers by ~9%. Corrects all five affected entries in both model_prices_and_context_window.json and the bundled model_prices_and_context_window_backup.json to match AWS's published GovCloud rates (per https://aws.amazon.com/bedrock/pricing/): - input_cost_per_token 3.3e-06 -> 3.6e-06 ($3.60/MTok) - output_cost_per_token 1.65e-05 -> 1.8e-05 ($18.00/MTok) - cache_creation_input_token_cost 4.125e-06 -> 4.5e-06 ($4.50/MTok) - cache_read_input_token_cost 3.3e-07 -> 3.6e-07 ($0.36/MTok) - cache_creation_input_token_cost_above_1hr (added) ($7.20/MTok) Affected entries: - bedrock/us-gov-east-1/anthropic.claude-sonnet-4-5-20250929-v1:0 - bedrock/us-gov-west-1/anthropic.claude-sonnet-4-5-20250929-v1:0 - bedrock/us-gov-east-1/claude-sonnet-4-5-20250929-v1:0 - bedrock/us-gov-west-1/claude-sonnet-4-5-20250929-v1:0 - us-gov.anthropic.claude-sonnet-4-5-20250929-v1:0 New regression test tests/test_litellm/test_bedrock_usgov_pricing.py pins the corrected rates across all five entries and asserts the 1.2x global-to-us-gov ratio invariant, matching the pattern used by the existing Vertex and EU/AU/JP Bedrock pricing tests.
This commit is contained in:
Filippo Mattia Menghi
parent
2c144fcfbc
commit
6cc1dad023
3 changed files with 139 additions and 40 deletions
|
|
@ -8724,15 +8724,16 @@
|
|||
"cache_creation_input_token_cost": 3.75e-07
|
||||
},
|
||||
"bedrock/us-gov-east-1/anthropic.claude-sonnet-4-5-20250929-v1:0": {
|
||||
"cache_creation_input_token_cost": 4.125e-06,
|
||||
"cache_read_input_token_cost": 3.3e-07,
|
||||
"input_cost_per_token": 3.3e-06,
|
||||
"cache_creation_input_token_cost": 4.5e-06,
|
||||
"cache_creation_input_token_cost_above_1hr": 7.2e-06,
|
||||
"cache_read_input_token_cost": 3.6e-07,
|
||||
"input_cost_per_token": 3.6e-06,
|
||||
"litellm_provider": "bedrock",
|
||||
"max_input_tokens": 200000,
|
||||
"max_output_tokens": 8192,
|
||||
"max_tokens": 8192,
|
||||
"mode": "chat",
|
||||
"output_cost_per_token": 1.65e-05,
|
||||
"output_cost_per_token": 1.8e-05,
|
||||
"supports_assistant_prefill": true,
|
||||
"supports_computer_use": true,
|
||||
"supports_function_calling": true,
|
||||
|
|
@ -8745,15 +8746,16 @@
|
|||
"supports_native_structured_output": true
|
||||
},
|
||||
"bedrock/us-gov-east-1/claude-sonnet-4-5-20250929-v1:0": {
|
||||
"cache_creation_input_token_cost": 4.125e-06,
|
||||
"cache_read_input_token_cost": 3.3e-07,
|
||||
"input_cost_per_token": 3.3e-06,
|
||||
"cache_creation_input_token_cost": 4.5e-06,
|
||||
"cache_creation_input_token_cost_above_1hr": 7.2e-06,
|
||||
"cache_read_input_token_cost": 3.6e-07,
|
||||
"input_cost_per_token": 3.6e-06,
|
||||
"litellm_provider": "bedrock",
|
||||
"max_input_tokens": 200000,
|
||||
"max_output_tokens": 8192,
|
||||
"max_tokens": 8192,
|
||||
"mode": "chat",
|
||||
"output_cost_per_token": 1.65e-05,
|
||||
"output_cost_per_token": 1.8e-05,
|
||||
"supports_assistant_prefill": true,
|
||||
"supports_computer_use": true,
|
||||
"supports_function_calling": true,
|
||||
|
|
@ -8897,15 +8899,16 @@
|
|||
"cache_creation_input_token_cost": 3.75e-07
|
||||
},
|
||||
"bedrock/us-gov-west-1/anthropic.claude-sonnet-4-5-20250929-v1:0": {
|
||||
"cache_creation_input_token_cost": 4.125e-06,
|
||||
"cache_read_input_token_cost": 3.3e-07,
|
||||
"input_cost_per_token": 3.3e-06,
|
||||
"cache_creation_input_token_cost": 4.5e-06,
|
||||
"cache_creation_input_token_cost_above_1hr": 7.2e-06,
|
||||
"cache_read_input_token_cost": 3.6e-07,
|
||||
"input_cost_per_token": 3.6e-06,
|
||||
"litellm_provider": "bedrock",
|
||||
"max_input_tokens": 200000,
|
||||
"max_output_tokens": 8192,
|
||||
"max_tokens": 8192,
|
||||
"mode": "chat",
|
||||
"output_cost_per_token": 1.65e-05,
|
||||
"output_cost_per_token": 1.8e-05,
|
||||
"supports_assistant_prefill": true,
|
||||
"supports_computer_use": true,
|
||||
"supports_function_calling": true,
|
||||
|
|
@ -8918,15 +8921,16 @@
|
|||
"supports_native_structured_output": true
|
||||
},
|
||||
"bedrock/us-gov-west-1/claude-sonnet-4-5-20250929-v1:0": {
|
||||
"cache_creation_input_token_cost": 4.125e-06,
|
||||
"cache_read_input_token_cost": 3.3e-07,
|
||||
"input_cost_per_token": 3.3e-06,
|
||||
"cache_creation_input_token_cost": 4.5e-06,
|
||||
"cache_creation_input_token_cost_above_1hr": 7.2e-06,
|
||||
"cache_read_input_token_cost": 3.6e-07,
|
||||
"input_cost_per_token": 3.6e-06,
|
||||
"litellm_provider": "bedrock",
|
||||
"max_input_tokens": 200000,
|
||||
"max_output_tokens": 8192,
|
||||
"max_tokens": 8192,
|
||||
"mode": "chat",
|
||||
"output_cost_per_token": 1.65e-05,
|
||||
"output_cost_per_token": 1.8e-05,
|
||||
"supports_assistant_prefill": true,
|
||||
"supports_computer_use": true,
|
||||
"supports_function_calling": true,
|
||||
|
|
@ -31364,9 +31368,10 @@
|
|||
"supports_native_structured_output": true
|
||||
},
|
||||
"us-gov.anthropic.claude-sonnet-4-5-20250929-v1:0": {
|
||||
"cache_creation_input_token_cost": 4.125e-06,
|
||||
"cache_read_input_token_cost": 3.3e-07,
|
||||
"input_cost_per_token": 3.3e-06,
|
||||
"cache_creation_input_token_cost": 4.5e-06,
|
||||
"cache_creation_input_token_cost_above_1hr": 7.2e-06,
|
||||
"cache_read_input_token_cost": 3.6e-07,
|
||||
"input_cost_per_token": 3.6e-06,
|
||||
"input_cost_per_token_above_200k_tokens": 6.6e-06,
|
||||
"output_cost_per_token_above_200k_tokens": 2.475e-05,
|
||||
"cache_creation_input_token_cost_above_200k_tokens": 8.25e-06,
|
||||
|
|
@ -31376,7 +31381,7 @@
|
|||
"max_output_tokens": 64000,
|
||||
"max_tokens": 64000,
|
||||
"mode": "chat",
|
||||
"output_cost_per_token": 1.65e-05,
|
||||
"output_cost_per_token": 1.8e-05,
|
||||
"supports_assistant_prefill": true,
|
||||
"supports_computer_use": true,
|
||||
"supports_function_calling": true,
|
||||
|
|
|
|||
|
|
@ -8737,15 +8737,16 @@
|
|||
"cache_creation_input_token_cost": 3.75e-07
|
||||
},
|
||||
"bedrock/us-gov-east-1/anthropic.claude-sonnet-4-5-20250929-v1:0": {
|
||||
"cache_creation_input_token_cost": 4.125e-06,
|
||||
"cache_read_input_token_cost": 3.3e-07,
|
||||
"input_cost_per_token": 3.3e-06,
|
||||
"cache_creation_input_token_cost": 4.5e-06,
|
||||
"cache_creation_input_token_cost_above_1hr": 7.2e-06,
|
||||
"cache_read_input_token_cost": 3.6e-07,
|
||||
"input_cost_per_token": 3.6e-06,
|
||||
"litellm_provider": "bedrock",
|
||||
"max_input_tokens": 200000,
|
||||
"max_output_tokens": 8192,
|
||||
"max_tokens": 8192,
|
||||
"mode": "chat",
|
||||
"output_cost_per_token": 1.65e-05,
|
||||
"output_cost_per_token": 1.8e-05,
|
||||
"supports_assistant_prefill": true,
|
||||
"supports_computer_use": true,
|
||||
"supports_function_calling": true,
|
||||
|
|
@ -8758,15 +8759,16 @@
|
|||
"supports_native_structured_output": true
|
||||
},
|
||||
"bedrock/us-gov-east-1/claude-sonnet-4-5-20250929-v1:0": {
|
||||
"cache_creation_input_token_cost": 4.125e-06,
|
||||
"cache_read_input_token_cost": 3.3e-07,
|
||||
"input_cost_per_token": 3.3e-06,
|
||||
"cache_creation_input_token_cost": 4.5e-06,
|
||||
"cache_creation_input_token_cost_above_1hr": 7.2e-06,
|
||||
"cache_read_input_token_cost": 3.6e-07,
|
||||
"input_cost_per_token": 3.6e-06,
|
||||
"litellm_provider": "bedrock",
|
||||
"max_input_tokens": 200000,
|
||||
"max_output_tokens": 8192,
|
||||
"max_tokens": 8192,
|
||||
"mode": "chat",
|
||||
"output_cost_per_token": 1.65e-05,
|
||||
"output_cost_per_token": 1.8e-05,
|
||||
"supports_assistant_prefill": true,
|
||||
"supports_computer_use": true,
|
||||
"supports_function_calling": true,
|
||||
|
|
@ -8910,15 +8912,16 @@
|
|||
"cache_creation_input_token_cost": 3.75e-07
|
||||
},
|
||||
"bedrock/us-gov-west-1/anthropic.claude-sonnet-4-5-20250929-v1:0": {
|
||||
"cache_creation_input_token_cost": 4.125e-06,
|
||||
"cache_read_input_token_cost": 3.3e-07,
|
||||
"input_cost_per_token": 3.3e-06,
|
||||
"cache_creation_input_token_cost": 4.5e-06,
|
||||
"cache_creation_input_token_cost_above_1hr": 7.2e-06,
|
||||
"cache_read_input_token_cost": 3.6e-07,
|
||||
"input_cost_per_token": 3.6e-06,
|
||||
"litellm_provider": "bedrock",
|
||||
"max_input_tokens": 200000,
|
||||
"max_output_tokens": 8192,
|
||||
"max_tokens": 8192,
|
||||
"mode": "chat",
|
||||
"output_cost_per_token": 1.65e-05,
|
||||
"output_cost_per_token": 1.8e-05,
|
||||
"supports_assistant_prefill": true,
|
||||
"supports_computer_use": true,
|
||||
"supports_function_calling": true,
|
||||
|
|
@ -8931,15 +8934,16 @@
|
|||
"supports_native_structured_output": true
|
||||
},
|
||||
"bedrock/us-gov-west-1/claude-sonnet-4-5-20250929-v1:0": {
|
||||
"cache_creation_input_token_cost": 4.125e-06,
|
||||
"cache_read_input_token_cost": 3.3e-07,
|
||||
"input_cost_per_token": 3.3e-06,
|
||||
"cache_creation_input_token_cost": 4.5e-06,
|
||||
"cache_creation_input_token_cost_above_1hr": 7.2e-06,
|
||||
"cache_read_input_token_cost": 3.6e-07,
|
||||
"input_cost_per_token": 3.6e-06,
|
||||
"litellm_provider": "bedrock",
|
||||
"max_input_tokens": 200000,
|
||||
"max_output_tokens": 8192,
|
||||
"max_tokens": 8192,
|
||||
"mode": "chat",
|
||||
"output_cost_per_token": 1.65e-05,
|
||||
"output_cost_per_token": 1.8e-05,
|
||||
"supports_assistant_prefill": true,
|
||||
"supports_computer_use": true,
|
||||
"supports_function_calling": true,
|
||||
|
|
@ -31420,9 +31424,10 @@
|
|||
"supports_native_structured_output": true
|
||||
},
|
||||
"us-gov.anthropic.claude-sonnet-4-5-20250929-v1:0": {
|
||||
"cache_creation_input_token_cost": 4.125e-06,
|
||||
"cache_read_input_token_cost": 3.3e-07,
|
||||
"input_cost_per_token": 3.3e-06,
|
||||
"cache_creation_input_token_cost": 4.5e-06,
|
||||
"cache_creation_input_token_cost_above_1hr": 7.2e-06,
|
||||
"cache_read_input_token_cost": 3.6e-07,
|
||||
"input_cost_per_token": 3.6e-06,
|
||||
"input_cost_per_token_above_200k_tokens": 6.6e-06,
|
||||
"output_cost_per_token_above_200k_tokens": 2.475e-05,
|
||||
"cache_creation_input_token_cost_above_200k_tokens": 8.25e-06,
|
||||
|
|
@ -31432,7 +31437,7 @@
|
|||
"max_output_tokens": 64000,
|
||||
"max_tokens": 64000,
|
||||
"mode": "chat",
|
||||
"output_cost_per_token": 1.65e-05,
|
||||
"output_cost_per_token": 1.8e-05,
|
||||
"supports_assistant_prefill": true,
|
||||
"supports_computer_use": true,
|
||||
"supports_function_calling": true,
|
||||
|
|
|
|||
89
tests/test_litellm/test_bedrock_usgov_pricing.py
Normal file
89
tests/test_litellm/test_bedrock_usgov_pricing.py
Normal file
|
|
@ -0,0 +1,89 @@
|
|||
"""
|
||||
Validate AWS GovCloud (Bedrock us-gov-*) Anthropic pricing entries.
|
||||
|
||||
AWS Bedrock pricing in GovCloud carries a +20% premium over the global
|
||||
Anthropic prices (not the +10% commercial-US premium). Until 2026-05-22
|
||||
these entries silently mirrored commercial US, undercharging customers
|
||||
by ~9%.
|
||||
|
||||
Source: https://aws.amazon.com/bedrock/pricing/
|
||||
|
||||
Sonnet 4.5 in us-gov-* (per million tokens):
|
||||
input = $3.60
|
||||
output = $18.00
|
||||
cache write 5m = $4.50
|
||||
cache write 1h = $7.20
|
||||
cache read = $0.36
|
||||
|
||||
Reference: https://github.com/BerriAI/litellm/issues/27120
|
||||
"""
|
||||
|
||||
import json
|
||||
import os
|
||||
|
||||
import pytest
|
||||
|
||||
|
||||
@pytest.fixture(scope="module")
|
||||
def model_data():
|
||||
json_path = os.path.join(
|
||||
os.path.dirname(__file__), "../../model_prices_and_context_window.json"
|
||||
)
|
||||
with open(json_path) as f:
|
||||
return json.load(f)
|
||||
|
||||
|
||||
SONNET_4_5_USGOV_KEYS = [
|
||||
"bedrock/us-gov-east-1/anthropic.claude-sonnet-4-5-20250929-v1:0",
|
||||
"bedrock/us-gov-west-1/anthropic.claude-sonnet-4-5-20250929-v1:0",
|
||||
"bedrock/us-gov-east-1/claude-sonnet-4-5-20250929-v1:0",
|
||||
"bedrock/us-gov-west-1/claude-sonnet-4-5-20250929-v1:0",
|
||||
"us-gov.anthropic.claude-sonnet-4-5-20250929-v1:0",
|
||||
]
|
||||
|
||||
|
||||
@pytest.mark.parametrize("model_key", SONNET_4_5_USGOV_KEYS)
|
||||
def test_usgov_sonnet_4_5_pricing(model_data, model_key):
|
||||
"""Each us-gov sonnet-4-5 entry must carry the +20%-over-global rates
|
||||
that AWS publishes on the GovCloud pricing page.
|
||||
"""
|
||||
assert model_key in model_data, f"Missing model entry: {model_key}"
|
||||
info = model_data[model_key]
|
||||
|
||||
assert info["input_cost_per_token"] == 3.6e-06, (
|
||||
f"{model_key}: input_cost_per_token should be $3.60/MTok "
|
||||
f"(got {info['input_cost_per_token']})"
|
||||
)
|
||||
assert (
|
||||
info["output_cost_per_token"] == 1.8e-05
|
||||
), f"{model_key}: output_cost_per_token should be $18.00/MTok"
|
||||
assert (
|
||||
info["cache_creation_input_token_cost"] == 4.5e-06
|
||||
), f"{model_key}: 5m cache write should be $4.50/MTok"
|
||||
assert (
|
||||
info["cache_creation_input_token_cost_above_1hr"] == 7.2e-06
|
||||
), f"{model_key}: 1h cache write should be $7.20/MTok"
|
||||
assert (
|
||||
info["cache_read_input_token_cost"] == 3.6e-07
|
||||
), f"{model_key}: cache read should be $0.36/MTok"
|
||||
|
||||
|
||||
def test_usgov_carries_20_percent_premium_over_global(model_data):
|
||||
"""The us-gov rates must equal 1.2x the global anthropic.* rates,
|
||||
matching AWS's documented GovCloud uplift.
|
||||
"""
|
||||
global_key = "anthropic.claude-sonnet-4-5-20250929-v1:0"
|
||||
usgov_key = "bedrock/us-gov-west-1/anthropic.claude-sonnet-4-5-20250929-v1:0"
|
||||
global_info = model_data[global_key]
|
||||
usgov_info = model_data[usgov_key]
|
||||
for field in (
|
||||
"input_cost_per_token",
|
||||
"output_cost_per_token",
|
||||
"cache_creation_input_token_cost",
|
||||
"cache_creation_input_token_cost_above_1hr",
|
||||
"cache_read_input_token_cost",
|
||||
):
|
||||
ratio = usgov_info[field] / global_info[field]
|
||||
assert (
|
||||
abs(ratio - 1.2) < 1e-9
|
||||
), f"{field}: us-gov / global ratio is {ratio}, expected 1.2"
|
||||
Loading…
Add table
Reference in a new issue