scriptoriumadmin/litellm
1
0
Fork 0
mirror of https://github.com/BerriAI/litellm.git synced 2026-09-27 01:22:18 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 82

feat(providers): add DeepSeek V4.1 Flash to Prism

Browse source
This commit is contained in:
rajitkhanna 2026-09-12 16:02:00 -07:00 • committed by mateo-berri
parent b87ffa2dc8
commit 40222e4815
4 changed files with 68 additions and 8 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

24
litellm/model_prices_and_context_window_backup.json
View file

@ -74783,6 +74783,30 @@
"supports_vision": false,
"supports_web_search": false
},
"prism/deepseek-v4.1-flash": {
"cache_read_input_token_cost": 7e-08,
"input_cost_per_token": 3e-07,
"litellm_provider": "prism",
"max_input_tokens": 1000000,
"max_output_tokens": 384000,
"max_tokens": 384000,
"mode": "chat",
"output_cost_per_token": 1.2e-06,
"source": "https://prisminference.com/pricing",
"supported_endpoints": [
"/v1/chat/completions",
"/v1/responses",
"/v1/messages"
],
"supports_function_calling": true,
"supports_native_streaming": true,
"supports_parallel_function_calling": true,
"supports_prompt_caching": true,
"supports_reasoning": true,
"supports_response_schema": true,
"supports_system_messages": true,
"supports_tool_choice": true
},
"prism/deepseek-v4-flash": {
"cache_read_input_token_cost": 7e-08,
"input_cost_per_token": 1.4e-07,

2
litellm/proxy/public_endpoints/provider_create_fields.json
View file

@ -2747,7 +2747,7 @@
"default_value": null
}
],
"default_model_placeholder": "prism/deepseek-v4-flash"
"default_model_placeholder": "prism/deepseek-v4.1-flash"
},
{
"provider": "RECRAFT",

24
model_prices_and_context_window.json
View file

@ -74783,6 +74783,30 @@
"supports_vision": false,
"supports_web_search": false
},
"prism/deepseek-v4.1-flash": {
"cache_read_input_token_cost": 7e-08,
"input_cost_per_token": 3e-07,
"litellm_provider": "prism",
"max_input_tokens": 1000000,
"max_output_tokens": 384000,
"max_tokens": 384000,
"mode": "chat",
"output_cost_per_token": 1.2e-06,
"source": "https://prisminference.com/pricing",
"supported_endpoints": [
"/v1/chat/completions",
"/v1/responses",
"/v1/messages"
],
"supports_function_calling": true,
"supports_native_streaming": true,
"supports_parallel_function_calling": true,
"supports_prompt_caching": true,
"supports_reasoning": true,
"supports_response_schema": true,
"supports_system_messages": true,
"supports_tool_choice": true
},
"prism/deepseek-v4-flash": {
"cache_read_input_token_cost": 7e-08,
"input_cost_per_token": 1.4e-07,

26
tests/test_litellm/llms/openai_like/test_prism_provider.py
View file

@ -41,22 +41,34 @@ def test_prism_provider_keeps_explicit_credentials(monkeypatch: pytest.MonkeyPat
assert api_base == "https://prism.internal.example/v1"
def test_prism_model_cost_and_capabilities():
@pytest.mark.parametrize(
("model", "input_cost", "output_cost", "max_output_tokens"),
[
("prism/deepseek-v4.1-flash", 0.30, 1.20, 384_000),
("prism/deepseek-v4-flash", 0.14, 0.28, 393_216),
],
)
def test_prism_model_cost_and_capabilities(
model: str,
input_cost: float,
output_cost: float,
max_output_tokens: int,
):
from litellm.cost_calculator import cost_per_token
prompt_cost, completion_cost = cost_per_token(
model="prism/deepseek-v4-flash",
model=model,
prompt_tokens=1_000_000,
completion_tokens=1_000_000,
custom_llm_provider="prism",
)
model_info = litellm.get_model_info("prism/deepseek-v4-flash")
model_info = litellm.get_model_info(model)
assert prompt_cost == pytest.approx(0.14)
assert completion_cost == pytest.approx(0.28)
assert prompt_cost == pytest.approx(input_cost)
assert completion_cost == pytest.approx(output_cost)
assert model_info["cache_read_input_token_cost"] == pytest.approx(7e-08)
assert model_info["max_input_tokens"] == 1_000_000
assert model_info["max_output_tokens"] == 393_216
assert model_info["max_output_tokens"] == max_output_tokens
assert model_info["supports_function_calling"] is True
assert model_info["supports_native_streaming"] is True
assert model_info["supports_reasoning"] is True
@ -70,7 +82,7 @@ def test_prism_is_available_in_add_model_form():
assert prism["provider"] == "PRISM"
assert prism["provider_display_name"] == "Prism"
assert prism["default_model_placeholder"] == "prism/deepseek-v4-flash"
assert prism["default_model_placeholder"] == "prism/deepseek-v4.1-flash"
assert {field["key"]: field["required"] for field in prism["credential_fields"]} == {
"api_base": False,
"api_key": True,