scriptoriumadmin/litellm
1
0
Fork 0
mirror of https://github.com/BerriAI/litellm.git synced 2026-08-28 05:25:59 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 36

fix(model_prices): sync gpt-5.6-sol bedrock rates, add gpt-5.6-cyber, fix claude 3 1h cache writes

Browse source 
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
This commit is contained in:
Devin AI 2026-08-25 13:26:48 +00:00
parent 31a67561ab
commit ab160fb953
4 changed files with 131 additions and 66 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

80
litellm/model_prices_and_context_window_backup.json
View file

@ -12282,7 +12282,7 @@
},
"claude-3-haiku-20240307": {
"cache_creation_input_token_cost": 3e-07,
"cache_creation_input_token_cost_above_1hr": 6e-06,
"cache_creation_input_token_cost_above_1hr": 5e-07,
"cache_read_input_token_cost": 3e-08,
"deprecation_date": "2026-04-20",
"input_cost_per_token": 2.5e-07,
@ -12301,7 +12301,7 @@
},
"claude-3-opus-20240229": {
"cache_creation_input_token_cost": 1.875e-05,
"cache_creation_input_token_cost_above_1hr": 6e-06,
"cache_creation_input_token_cost_above_1hr": 3e-05,
"cache_read_input_token_cost": 1.5e-06,
"deprecation_date": "2026-01-05",
"input_cost_per_token": 1.5e-05,
@ -49007,14 +49007,14 @@
"supports_tool_choice": true
},
"bedrock_mantle/openai.gpt-5.6-sol": {
"input_cost_per_token": 5.5e-06,
"input_cost_per_token_above_272k_tokens": 1.1e-05,
"cache_creation_input_token_cost": 6.875e-06,
"cache_creation_input_token_cost_above_272k_tokens": 1.375e-05,
"cache_read_input_token_cost": 5.5e-07,
"cache_read_input_token_cost_above_272k_tokens": 1.1e-06,
"output_cost_per_token": 3.3e-05,
"output_cost_per_token_above_272k_tokens": 4.95e-05,
"input_cost_per_token": 4.4e-06,
"input_cost_per_token_above_272k_tokens": 8.8e-06,
"cache_creation_input_token_cost": 5.5e-06,
"cache_creation_input_token_cost_above_272k_tokens": 1.1e-05,
"cache_read_input_token_cost": 4.4e-07,
"cache_read_input_token_cost_above_272k_tokens": 8.8e-07,
"output_cost_per_token": 2.2e-05,
"output_cost_per_token_above_272k_tokens": 3.3e-05,
"litellm_provider": "bedrock_mantle",
"max_input_tokens": 1000000,
"max_output_tokens": 128000,
@ -49070,6 +49070,34 @@
"supports_tool_choice": true,
"supports_vision": true
},
"bedrock_mantle/openai.gpt-5.6-cyber": {
"input_cost_per_token": 1.375e-05,
"cache_creation_input_token_cost": 1.71875e-05,
"cache_read_input_token_cost": 1.375e-06,
"output_cost_per_token": 8.25e-05,
"litellm_provider": "bedrock_mantle",
"max_input_tokens": 272000,
"max_output_tokens": 128000,
"max_tokens": 128000,
"mode": "responses",
"use_openai_responses_path": true,
"supported_endpoints": [
"/v1/responses"
],
"supported_modalities": [
"text",
"image"
],
"supported_output_modalities": [
"text"
],
"supports_function_calling": true,
"supports_prompt_caching": true,
"supports_reasoning": true,
"supports_response_schema": true,
"supports_tool_choice": true,
"supports_vision": true
},
"bedrock_mantle/openai.gpt-5.6-luna": {
"input_cost_per_token": 2.2e-07,
"input_cost_per_token_above_272k_tokens": 4.4e-07,
@ -49103,14 +49131,14 @@
"supports_vision": true
},
"us.openai.gpt-5.6-sol": {
"input_cost_per_token": 5.5e-06,
"input_cost_per_token_above_272k_tokens": 1.1e-05,
"cache_creation_input_token_cost": 6.875e-06,
"cache_creation_input_token_cost_above_272k_tokens": 1.375e-05,
"cache_read_input_token_cost": 5.5e-07,
"cache_read_input_token_cost_above_272k_tokens": 1.1e-06,
"output_cost_per_token": 3.3e-05,
"output_cost_per_token_above_272k_tokens": 4.95e-05,
"input_cost_per_token": 4.4e-06,
"input_cost_per_token_above_272k_tokens": 8.8e-06,
"cache_creation_input_token_cost": 5.5e-06,
"cache_creation_input_token_cost_above_272k_tokens": 1.1e-05,
"cache_read_input_token_cost": 4.4e-07,
"cache_read_input_token_cost_above_272k_tokens": 8.8e-07,
"output_cost_per_token": 2.2e-05,
"output_cost_per_token_above_272k_tokens": 3.3e-05,
"litellm_provider": "bedrock_converse",
"max_input_tokens": 1000000,
"max_output_tokens": 128000,
@ -49128,14 +49156,14 @@
"supports_vision": true
},
"global.openai.gpt-5.6-sol": {
"input_cost_per_token": 5e-06,
"input_cost_per_token_above_272k_tokens": 1e-05,
"cache_creation_input_token_cost": 6.25e-06,
"cache_creation_input_token_cost_above_272k_tokens": 1.25e-05,
"cache_read_input_token_cost": 5e-07,
"cache_read_input_token_cost_above_272k_tokens": 1e-06,
"output_cost_per_token": 3e-05,
"output_cost_per_token_above_272k_tokens": 4.5e-05,
"input_cost_per_token": 4e-06,
"input_cost_per_token_above_272k_tokens": 8e-06,
"cache_creation_input_token_cost": 5e-06,
"cache_creation_input_token_cost_above_272k_tokens": 1e-05,
"cache_read_input_token_cost": 4e-07,
"cache_read_input_token_cost_above_272k_tokens": 8e-07,
"output_cost_per_token": 2e-05,
"output_cost_per_token_above_272k_tokens": 3e-05,
"litellm_provider": "bedrock_converse",
"max_input_tokens": 1000000,
"max_output_tokens": 128000,

80
model_prices_and_context_window.json
View file

@ -12282,7 +12282,7 @@
},
"claude-3-haiku-20240307": {
"cache_creation_input_token_cost": 3e-07,
"cache_creation_input_token_cost_above_1hr": 6e-06,
"cache_creation_input_token_cost_above_1hr": 5e-07,
"cache_read_input_token_cost": 3e-08,
"deprecation_date": "2026-04-20",
"input_cost_per_token": 2.5e-07,
@ -12301,7 +12301,7 @@
},
"claude-3-opus-20240229": {
"cache_creation_input_token_cost": 1.875e-05,
"cache_creation_input_token_cost_above_1hr": 6e-06,
"cache_creation_input_token_cost_above_1hr": 3e-05,
"cache_read_input_token_cost": 1.5e-06,
"deprecation_date": "2026-01-05",
"input_cost_per_token": 1.5e-05,
@ -49007,14 +49007,14 @@
"supports_tool_choice": true
},
"bedrock_mantle/openai.gpt-5.6-sol": {
"input_cost_per_token": 5.5e-06,
"input_cost_per_token_above_272k_tokens": 1.1e-05,
"cache_creation_input_token_cost": 6.875e-06,
"cache_creation_input_token_cost_above_272k_tokens": 1.375e-05,
"cache_read_input_token_cost": 5.5e-07,
"cache_read_input_token_cost_above_272k_tokens": 1.1e-06,
"output_cost_per_token": 3.3e-05,
"output_cost_per_token_above_272k_tokens": 4.95e-05,
"input_cost_per_token": 4.4e-06,
"input_cost_per_token_above_272k_tokens": 8.8e-06,
"cache_creation_input_token_cost": 5.5e-06,
"cache_creation_input_token_cost_above_272k_tokens": 1.1e-05,
"cache_read_input_token_cost": 4.4e-07,
"cache_read_input_token_cost_above_272k_tokens": 8.8e-07,
"output_cost_per_token": 2.2e-05,
"output_cost_per_token_above_272k_tokens": 3.3e-05,
"litellm_provider": "bedrock_mantle",
"max_input_tokens": 1000000,
"max_output_tokens": 128000,
@ -49070,6 +49070,34 @@
"supports_tool_choice": true,
"supports_vision": true
},
"bedrock_mantle/openai.gpt-5.6-cyber": {
"input_cost_per_token": 1.375e-05,
"cache_creation_input_token_cost": 1.71875e-05,
"cache_read_input_token_cost": 1.375e-06,
"output_cost_per_token": 8.25e-05,
"litellm_provider": "bedrock_mantle",
"max_input_tokens": 272000,
"max_output_tokens": 128000,
"max_tokens": 128000,
"mode": "responses",
"use_openai_responses_path": true,
"supported_endpoints": [
"/v1/responses"
],
"supported_modalities": [
"text",
"image"
],
"supported_output_modalities": [
"text"
],
"supports_function_calling": true,
"supports_prompt_caching": true,
"supports_reasoning": true,
"supports_response_schema": true,
"supports_tool_choice": true,
"supports_vision": true
},
"bedrock_mantle/openai.gpt-5.6-luna": {
"input_cost_per_token": 2.2e-07,
"input_cost_per_token_above_272k_tokens": 4.4e-07,
@ -49103,14 +49131,14 @@
"supports_vision": true
},
"us.openai.gpt-5.6-sol": {
"input_cost_per_token": 5.5e-06,
"input_cost_per_token_above_272k_tokens": 1.1e-05,
"cache_creation_input_token_cost": 6.875e-06,
"cache_creation_input_token_cost_above_272k_tokens": 1.375e-05,
"cache_read_input_token_cost": 5.5e-07,
"cache_read_input_token_cost_above_272k_tokens": 1.1e-06,
"output_cost_per_token": 3.3e-05,
"output_cost_per_token_above_272k_tokens": 4.95e-05,
"input_cost_per_token": 4.4e-06,
"input_cost_per_token_above_272k_tokens": 8.8e-06,
"cache_creation_input_token_cost": 5.5e-06,
"cache_creation_input_token_cost_above_272k_tokens": 1.1e-05,
"cache_read_input_token_cost": 4.4e-07,
"cache_read_input_token_cost_above_272k_tokens": 8.8e-07,
"output_cost_per_token": 2.2e-05,
"output_cost_per_token_above_272k_tokens": 3.3e-05,
"litellm_provider": "bedrock_converse",
"max_input_tokens": 1000000,
"max_output_tokens": 128000,
@ -49128,14 +49156,14 @@
"supports_vision": true
},
"global.openai.gpt-5.6-sol": {
"input_cost_per_token": 5e-06,
"input_cost_per_token_above_272k_tokens": 1e-05,
"cache_creation_input_token_cost": 6.25e-06,
"cache_creation_input_token_cost_above_272k_tokens": 1.25e-05,
"cache_read_input_token_cost": 5e-07,
"cache_read_input_token_cost_above_272k_tokens": 1e-06,
"output_cost_per_token": 3e-05,
"output_cost_per_token_above_272k_tokens": 4.5e-05,
"input_cost_per_token": 4e-06,
"input_cost_per_token_above_272k_tokens": 8e-06,
"cache_creation_input_token_cost": 5e-06,
"cache_creation_input_token_cost_above_272k_tokens": 1e-05,
"cache_read_input_token_cost": 4e-07,
"cache_read_input_token_cost_above_272k_tokens": 8e-07,
"output_cost_per_token": 2e-05,
"output_cost_per_token_above_272k_tokens": 3e-05,
"litellm_provider": "bedrock_converse",
"max_input_tokens": 1000000,
"max_output_tokens": 128000,

24
tests/test_litellm/llms/bedrock/test_cross_region_inference_profile_mapping.py
View file

@ -59,17 +59,17 @@ class GptProfile(NamedTuple):
GPT_5_6_PROFILES = [
GptProfile(
model_id="us.openai.gpt-5.6-sol",
input_cost=5.5e-06, input_cost_above_272k=1.1e-05,
cache_write=6.875e-06, cache_write_above_272k=1.375e-05,
cache_read=5.5e-07, cache_read_above_272k=1.1e-06,
output_cost=3.3e-05, output_cost_above_272k=4.95e-05,
input_cost=4.4e-06, input_cost_above_272k=8.8e-06,
cache_write=5.5e-06, cache_write_above_272k=1.1e-05,
cache_read=4.4e-07, cache_read_above_272k=8.8e-07,
output_cost=2.2e-05, output_cost_above_272k=3.3e-05,
),
GptProfile(
model_id="global.openai.gpt-5.6-sol",
input_cost=5e-06, input_cost_above_272k=1e-05,
cache_write=6.25e-06, cache_write_above_272k=1.25e-05,
cache_read=5e-07, cache_read_above_272k=1e-06,
output_cost=3e-05, output_cost_above_272k=4.5e-05,
input_cost=4e-06, input_cost_above_272k=8e-06,
cache_write=5e-06, cache_write_above_272k=1e-05,
cache_read=4e-07, cache_read_above_272k=8e-07,
output_cost=2e-05, output_cost_above_272k=3e-05,
),
GptProfile(
model_id="us.openai.gpt-5.6-terra",
@ -221,7 +221,7 @@ def test_bedrock_gpt_5_6_above_272k_tier_applies_to_cost(local_model_cost_map):
custom_llm_provider="bedrock",
)
assert cost == pytest.approx((300000 * 1.1e-05) + (1000 * 4.95e-05), rel=1e-9)
assert cost == pytest.approx((300000 * 8.8e-06) + (1000 * 3.3e-05), rel=1e-9)
def test_bedrock_gpt_5_6_bills_cache_read_tokens(local_model_cost_map):
@ -241,10 +241,10 @@ def test_bedrock_gpt_5_6_bills_cache_read_tokens(local_model_cost_map):
custom_llm_provider="bedrock",
)
expected = (2 * 5.5e-06) + (15609 * 5.5e-07) + (5 * 3.3e-05)
expected = (2 * 4.4e-06) + (15609 * 4.4e-07) + (5 * 2.2e-05)
assert cost == pytest.approx(expected, rel=1e-9)
# Without cache_read_input_token_cost the cached prefix bills at zero.
assert cost > (15611 * 5.5e-06) * 0.1
assert cost > (15611 * 4.4e-06) * 0.1
def test_bedrock_gpt_5_6_bills_cache_write_tokens(local_model_cost_map):
@ -263,7 +263,7 @@ def test_bedrock_gpt_5_6_bills_cache_write_tokens(local_model_cost_map):
custom_llm_provider="bedrock",
)
expected = (2 * 5.5e-06) + (15609 * 6.875e-06) + (5 * 3.3e-05)
expected = (2 * 4.4e-06) + (15609 * 5.5e-06) + (5 * 2.2e-05)
assert cost == pytest.approx(expected, rel=1e-9)

13
tests/test_litellm/llms/bedrock_mantle/test_bedrock_mantle_responses_transformation.py
View file

@ -1506,10 +1506,19 @@ class TestBedrockMantleResponsesPricing:
assert info["cache_read_input_token_cost"] == pytest.approx(2.75e-07)
assert info["max_input_tokens"] == 272000
def test_gpt_5_6_cyber_pricing_and_mode(self, local_cost_map):
info = litellm.get_model_info("bedrock_mantle/openai.gpt-5.6-cyber")
assert info["mode"] == "responses"
assert info["input_cost_per_token"] == pytest.approx(1.375e-05)
assert info["cache_creation_input_token_cost"] == pytest.approx(1.71875e-05)
assert info["cache_read_input_token_cost"] == pytest.approx(1.375e-06)
assert info["output_cost_per_token"] == pytest.approx(8.25e-05)
assert info["max_input_tokens"] == 272000
@pytest.mark.parametrize(
"model, input_cost, cache_creation_cost, cache_read_cost, output_cost",
[
("openai.gpt-5.6-sol", 5.5e-06, 6.875e-06, 5.5e-07, 3.3e-05),
("openai.gpt-5.6-sol", 4.4e-06, 5.5e-06, 4.4e-07, 2.2e-05),
("openai.gpt-5.6-terra", 2.2e-06, 2.75e-06, 2.2e-07, 1.32e-05),
("openai.gpt-5.6-luna", 2.2e-07, 2.75e-07, 2.2e-08, 1.32e-06),
],
@ -1532,7 +1541,7 @@ class TestBedrockMantleResponsesPricing:
@pytest.mark.parametrize(
"model, input_cost, output_cost",
[
("openai.gpt-5.6-sol", 5.5e-06, 3.3e-05),
("openai.gpt-5.6-sol", 4.4e-06, 2.2e-05),
("openai.gpt-5.6-terra", 2.2e-06, 1.32e-05),
("openai.gpt-5.6-luna", 2.2e-07, 1.32e-06),
],