scriptoriumadmin/litellm
1
0
Fork 0
mirror of https://github.com/BerriAI/litellm.git synced 2026-09-07 08:26:10 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 36

fix(model_prices): align fireworks deepseek-v4-flash-0731 twin cost entries at published pricing

Browse source
This commit is contained in:
mateo-berri 2026-08-29 13:29:35 -07:00
parent e0ed0a4c7a
commit 71d1237283
3 changed files with 44 additions and 6 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

6
litellm/model_prices_and_context_window_backup.json
View file

@ -51947,14 +51947,14 @@
"supports_vision": true
},
"fireworks_ai/deepseek-v4-flash-0731": {
"cache_read_input_token_cost": 2.8e-08,
"input_cost_per_token": 1.4e-07,
"cache_read_input_token_cost": 7e-09,
"input_cost_per_token": 2.2e-07,
"litellm_provider": "fireworks_ai",
"max_input_tokens": 1048576,
"max_output_tokens": 131072,
"max_tokens": 131072,
"mode": "chat",
"output_cost_per_token": 2.8e-07,
"output_cost_per_token": 6.6e-07,
"source": "https://docs.fireworks.ai/serverless/pricing",
"supports_function_calling": true,
"supports_reasoning": true,

6
model_prices_and_context_window.json
View file

@ -51947,14 +51947,14 @@
"supports_vision": true
},
"fireworks_ai/deepseek-v4-flash-0731": {
"cache_read_input_token_cost": 2.8e-08,
"input_cost_per_token": 1.4e-07,
"cache_read_input_token_cost": 7e-09,
"input_cost_per_token": 2.2e-07,
"litellm_provider": "fireworks_ai",
"max_input_tokens": 1048576,
"max_output_tokens": 131072,
"max_tokens": 131072,
"mode": "chat",
"output_cost_per_token": 2.8e-07,
"output_cost_per_token": 6.6e-07,
"source": "https://docs.fireworks.ai/serverless/pricing",
"supports_function_calling": true,
"supports_reasoning": true,

38
tests/test_litellm/test_fireworks_serverless_model_costs.py
View file

@ -84,3 +84,41 @@ def test_bare_fireworks_ids_resolve_through_prefixed_entries():
assert info["output_cost_per_token"] == pytest.approx(expected["output_cost_per_token"])
assert info["max_input_tokens"] == expected["max_input_tokens"]
assert info["max_output_tokens"] == expected["max_output_tokens"]
TWIN_PINNED_PRICES = {
"deepseek-v4-flash-0731": {
"input_cost_per_token": 2.2e-07,
"cache_read_input_token_cost": 7e-09,
"output_cost_per_token": 6.6e-07,
},
}
def test_deepseek_v4_flash_0731_twins_pin_published_pricing(model_data):
"""Both 0731 entries carry the price published at docs.fireworks.ai/serverless/pricing."""
for bare_suffix, expected in TWIN_PINNED_PRICES.items():
for key in (
f"fireworks_ai/{bare_suffix}",
f"fireworks_ai/accounts/fireworks/models/{bare_suffix}",
):
entry = model_data[key]
for field, value in expected.items():
assert entry[field] == pytest.approx(value), f"{key}.{field}"
def test_fireworks_account_prefixed_twins_agree_on_price(model_data):
"""Every accounts/fireworks/models/X entry prices identically to its bare fireworks_ai/X twin."""
prefix = "fireworks_ai/accounts/fireworks/models/"
pairs_checked = 0
for key, entry in model_data.items():
if not key.startswith(prefix):
continue
bare_key = f"fireworks_ai/{key[len(prefix):]}"
bare_entry = model_data.get(bare_key)
if bare_entry is None:
continue
pairs_checked += 1
for field in sorted({f for f in (*entry, *bare_entry) if "cost" in f}):
assert entry.get(field) == bare_entry.get(field), f"{key} vs {bare_key}: {field}"
assert pairs_checked >= 20