scriptoriumadmin/litellm
1
0
Fork 0
mirror of https://github.com/BerriAI/litellm.git synced 2026-09-12 23:01:41 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 89

Add promo pricing for gpt 5.6 sol

Browse source
This commit is contained in:
Liam McDonald 2026-08-24 11:53:29 -07:00
parent a626170c89
commit 6dbee8c449
4 changed files with 81 additions and 62 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

48
litellm/model_prices_and_context_window_backup.json
View file

@ -48833,14 +48833,14 @@
"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,
@ -48929,14 +48929,14 @@
"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,
@ -48954,14 +48954,14 @@
"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,

48
model_prices_and_context_window.json
View file

@ -48833,14 +48833,14 @@
"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,
@ -48929,14 +48929,14 @@
"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,
@ -48954,14 +48954,14 @@
"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,

32
tests/test_litellm/llms/bedrock/test_cross_region_inference_profile_mapping.py
View file

@ -57,19 +57,27 @@ class GptProfile(NamedTuple):
GPT_5_6_PROFILES = [
# AWS Sol promo at least through 2026-11-21. After that date restore sticker
# rates here and in both cost maps:
# us.openai.gpt-5.6-sol (Geo, same as Mantle In-Region):
# 5.5e-06 / 1.1e-05 input, 6.875e-06 / 1.375e-05 cache write,
# 5.5e-07 / 1.1e-06 cache read, 3.3e-05 / 4.95e-05 output
# global.openai.gpt-5.6-sol (Global CRIS):
# 5e-06 / 1e-05 input, 6.25e-06 / 1.25e-05 cache write,
# 5e-07 / 1e-06 cache read, 3e-05 / 4.5e-05 output
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 +229,7 @@ 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 +249,10 @@ 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 +271,7 @@ 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)

15
tests/test_litellm/llms/bedrock_mantle/test_bedrock_mantle_responses_transformation.py
View file

@ -1509,7 +1509,18 @@ class TestBedrockMantleResponsesPricing:
@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),
# AWS Sol promo (In-Region / Mantle) at least through 2026-11-21.
# After that date restore sticker rates here and in both cost maps
# (bedrock_mantle/openai.gpt-5.6-sol, and Geo us.openai.gpt-5.6-sol):
# input_cost_per_token: 5.5e-06
# cache_creation_input_token_cost: 6.875e-06
# cache_read_input_token_cost: 5.5e-07
# output_cost_per_token: 3.3e-05
# input_cost_per_token_above_272k_tokens: 1.1e-05
# cache_creation_input_token_cost_above_272k_tokens: 1.375e-05
# cache_read_input_token_cost_above_272k_tokens: 1.1e-06
# output_cost_per_token_above_272k_tokens: 4.95e-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 +1543,7 @@ 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),
],