Skip to content

The most powerful and modular diffusion model GUI, api and backend with a graph/nodes interface. The fastest local inference engine in the world.

[Partner Nodes] feat(credits): respect "X-Comfy-Credits-Used" header from the comfy-api (#15091)

Alexander Piskuncd0eddaf1616
diff --git a/comfy_api_nodes/nodes_anthropic.py b/comfy_api_nodes/nodes_anthropic.py
index 011c3f0cf..76c611b93 100644
--- a/comfy_api_nodes/nodes_anthropic.py
+++ b/comfy_api_nodes/nodes_anthropic.py
@@ -106,49 +106,6 @@ def _claude_model_inputs(model_label: str):
return inputs
-def _model_price_per_million(model: str) -> tuple[float, float] | None:
- """Return (input_per_1M, output_per_1M) USD for a Claude model, or None if unknown."""
- if "fable-5" in model:
- return 14.30, 71.50
- if "opus-5" in model or "opus-4-8" in model:
- return 7.15, 35.75
- if "sonnet-5" in model:
- return 2.86, 14.30
- if "opus-4-7" in model or "opus-4-6" in model or "opus-4-5" in model:
- return 5.0, 25.0
- if "sonnet-4" in model:
- return 3.0, 15.0
- if "haiku-4-5" in model:
- return 1.0, 5.0
- return None
-
-
-def calculate_tokens_price(response: AnthropicMessagesResponse) -> float | None:
- """Compute approximate USD price from response usage. Server-side billing is authoritative."""
- if not response.usage or not response.model:
- return None
- rates = _model_price_per_million(response.model)
- if rates is None:
- return None
- input_rate, output_rate = rates
- input_tokens = response.usage.input_tokens or 0
- output_tokens = response.usage.output_tokens or 0
- cache_read = response.usage.cache_read_input_tokens or 0
- cache_5m = 0
- cache_1h = 0
- if response.usage.cache_creation:
- cache_5m = response.usage.cache_creation.ephemeral_5m_input_tokens or 0
- cache_1h = response.usage.cache_creation.ephemeral_1h_input_tokens or 0
- total = (
- input_tokens * input_rate
- + output_tokens * output_rate
- + cache_read * input_rate * 0.1
- + cache_5m * input_rate * 1.25
- + cache_1h * input_rate * 2.0
- )
- return total / 1_000_000.0
-
-
def _get_text_from_response(response: AnthropicMessagesResponse) -> str:
if not response.content:
return ""
@@ -344,7 +301,6 @@ class ClaudeNode(IO.ComfyNode):
thinking=thinking_cfg,
output_config=output_cfg,
),
- price_extractor=calculate_tokens_price,
)
if response.stop_reason == "refusal":
raise ValueError(
diff --git a/comfy_api_nodes/nodes_bytedance_llm.py b/comfy_api_nodes/nodes_bytedance_llm.py
index cb41defa0..0403e0c1f 100644
--- a/comfy_api_nodes/nodes_bytedance_llm.py
+++ b/comfy_api_nodes/nodes_bytedance_llm.py
@@ -34,13 +34,6 @@ SEED_MODELS: dict[str, str] = {
"Seed 2.0 Mini": "seed-2-0-mini-260215",
}
-# USD per 1M tokens: (input, cache_hit_input, output)
-_SEED_PRICES_PER_MILLION: dict[str, tuple[float, float, float]] = {
- "seed-2-0-pro-260328": (0.50, 0.10, 3.00),
- "seed-2-0-lite-260228": (0.25, 0.05, 2.00),
- "seed-2-0-mini-260215": (0.10, 0.02, 0.40),
-}
-
def _seed_model_inputs(max_images: int = SEED_MAX_IMAGES, max_videos: int = SEED_MAX_VIDEOS):
return [
@@ -74,24 +67,6 @@ def _seed_model_inputs(max_images: int = SEED_MAX_IMAGES, max_videos: int = SEED
]
-def _calculate_price(model_id: str, response: BytePlusResponseObject) -> float | None:
- """Compute approximate USD price from response usage."""
- if not response.usage:
- return None
- rates = _SEED_PRICES_PER_MILLION.get(model_id)
- if rates is None:
- return None
- input_rate, cache_hit_rate, output_rate = rates
- input_tokens = response.usage.input_tokens or 0
- output_tokens = response.usage.output_tokens or 0
- cached = 0
- if response.usage.input_tokens_details:
- cached = response.usage.input_tokens_details.cached_tokens or 0
- fresh_input = max(0, input_tokens - cached)
- total = fresh_input * input_rate + cached * cache_hit_rate + output_tokens * output_rate
- return total / 1_000_000.0
-
-
def _get_text_from_response(response: BytePlusResponseObject) -> str:
"""Extract concatenated text from all assistant message output_text blocks."""
if not response.output:
@@ -251,7 +226,6 @@ class ByteDanceSeedNode(IO.ComfyNode):
store=False,
stream=False,
),
- price_extractor=lambda r: _calculate_price(model_id, r),
)
if response.error:
raise ValueError(f"Seed API error ({response.error.code}): {response.error.message}")
diff --git a/comfy_api_nodes/nodes_gemini.py b/comfy_api_nodes/nodes_gemini.py
index 47d028c6c..fd9ff04a8 100644
--- a/comfy_api_nodes/nodes_gemini.py
+++ b/comfy_api_nodes/nodes_gemini.py
@@ -35,7 +35,6 @@ from comfy_api_nodes.apis.gemini import (
GeminiSystemInstructionContent,
GeminiTextPart,
GeminiThinkingConfig,
- Modality,
)
from comfy_api_nodes.util import (
ApiEndpoint,
@@ -238,60 +237,6 @@ async def get_image_from_response(response: GeminiGenerateContentResponse, thoug
return torch.cat(image_tensors, dim=0)
-def calculate_tokens_price(response: GeminiGenerateContentResponse) -> float | None:
- if not response.modelVersion:
- return None
- # Define prices (Cost per 1,000,000 tokens), see https://cloud.google.com/vertex-ai/generative-ai/pricing
- if response.modelVersion == "gemini-2.5-pro":
- input_tokens_price = 1.25
- output_text_tokens_price = 10.0
- output_image_tokens_price = 0.0
- elif response.modelVersion == "gemini-2.5-flash":
- input_tokens_price = 0.30
- output_text_tokens_price = 2.50
- output_image_tokens_price = 0.0
- elif response.modelVersion == "gemini-2.5-flash-image":
- input_tokens_price = 0.30
- output_text_tokens_price = 2.50
- output_image_tokens_price = 30.0
- elif response.modelVersion in ("gemini-3-pro-preview", "gemini-3.1-pro-preview"):
- input_tokens_price = 2
- output_text_tokens_price = 12.0
- output_image_tokens_price = 0.0
- elif response.modelVersion in ("gemini-3.1-flash-lite-preview", "gemini-3.1-flash-lite"):
- input_tokens_price = 0.25
- output_text_tokens_price = 1.50
- output_image_tokens_price = 0.0
- elif response.modelVersion == "gemini-3.5-flash":
- input_tokens_price = 1.50
- output_text_tokens_price = 9.0
- output_image_tokens_price = 0.0
- elif response.modelVersion in ("gemini-3-pro-image-preview", "gemini-3-pro-image"):
- input_tokens_price = 2
- output_text_tokens_price = 12.0
- output_image_tokens_price = 120.0
- elif response.modelVersion in ("gemini-3.1-flash-image-preview", "gemini-3.1-flash-image"):
- input_tokens_price = 0.5
- output_text_tokens_price = 3.0
- output_image_tokens_price = 60.0
- elif response.modelVersion == "gemini-3.1-flash-lite-image":
- input_tokens_price = 0.25
- output_text_tokens_price = 1.50
- output_image_tokens_price = 30.0
- else:
- return None
- final_price = response.usageMetadata.promptTokenCount * input_tokens_price
- if response.usageMetadata.candidatesTokensDetails:
- for i in response.usageMetadata.candidatesTokensDetails:
- if i.modality == Modality.IMAGE:
- final_price += output_image_tokens_price * i.tokenCount # for Nano Banana models
- else:
- final_price += output_text_tokens_price * i.tokenCount
- if response.usageMetadata.thoughtsTokenCount:
- final_price += output_text_tokens_price * response.usageMetadata.thoughtsTokenCount
- return final_price / 1_000_000.0
-
-
def get_text_from_interaction(interaction: GeminiInteraction) -> str:
"""Extract and concatenate all model output text from an Interactions API response."""
texts = []
@@ -326,24 +271,6 @@ async def get_video_from_interaction(
)
-def calculate_interaction_tokens_price(interaction: GeminiInteraction) -> float | None:
- if interaction.usage is None:
- return None
- input_tokens_price = 1.5
- output_tokens_prices = {"text": 9.0, "video": 17.5}
- thoughts_tokens_price = 9.0
- final_price = 0.0
- for i in interaction.usage.input_tokens_by_modality or []:
- if i.tokens:
- final_price += input_tokens_price * i.tokens
- for i in interaction.usage.output_tokens_by_modality or []:
- if i.tokens and i.modality in output_tokens_prices:
- final_price += output_tokens_prices[i.modality] * i.tokens
- if interaction.usage.total_thought_tokens:
- final_price += thoughts_tokens_price * interaction.usage.total_thought_tokens
- return final_price / 1_000_000.0
-
-
def create_video_parts(video_input: Input.Video) -> list[GeminiPart]:
"""Convert a single video input to Gemini API compatible parts (inline MP4/H.264)."""
base_64_string = video_to_base64_string(
@@ -657,7 +584,6 @@ class GeminiNode(IO.ComfyNode):
systemInstruction=gemini_system_prompt,
),
response_model=GeminiGenerateContentResponse,
- price_extractor=calculate_tokens_price,
)
output_text = get_text_from_response(response)
@@ -872,7 +798,6 @@ class GeminiNodeV2(IO.ComfyNode):
systemInstruction=gemini_system_prompt,
),
response_model=GeminiGenerateContentResponse,
- price_extractor=calculate_tokens_price,
)
output_text = get_text_from_response(response)
@@ -1085,7 +1010,6 @@ class GeminiImage(IO.ComfyNode):
systemInstruction=gemini_system_prompt,
),
response_model=GeminiGenerateContentResponse,
- price_extractor=calculate_tokens_price,
)
return IO.NodeOutput(await get_image_from_response(response), get_text_from_response(response))
@@ -1225,7 +1149,6 @@ class GeminiImage2(IO.ComfyNode):
systemInstruction=gemini_system_prompt,
),
response_model=GeminiGenerateContentResponse,
- price_extractor=calculate_tokens_price,
)
return IO.NodeOutput(await get_image_from_response(response), get_text_from_response(response))
@@ -1385,7 +1308,6 @@ class GeminiNanoBanana2(IO.ComfyNode):
systemInstruction=gemini_system_prompt,
),
response_model=GeminiGenerateContentResponse,
- price_extractor=calculate_tokens_price,
)
return IO.NodeOutput(
await get_image_from_response(response),
@@ -1610,7 +1532,6 @@ class GeminiNanoBanana2V2(IO.ComfyNode):
systemInstruction=gemini_system_prompt,
),
response_model=GeminiGenerateContentResponse,
- price_extractor=calculate_tokens_price,
)
return IO.NodeOutput(
await get_image_from_response(response),
@@ -1762,7 +1683,6 @@ class GeminiVideoOmni(IO.ComfyNode):
),
),
response_model=GeminiInteraction,
- price_extractor=calculate_interaction_tokens_price,
)
if interaction.status != "completed":
model_message = get_text_from_interaction(interaction).strip()
diff --git a/comfy_api_nodes/nodes_grok.py b/comfy_api_nodes/nodes_grok.py
index dc484536e..a95b35917 100644
--- a/comfy_api_nodes/nodes_grok.py
+++ b/comfy_api_nodes/nodes_grok.py
@@ -155,7 +155,6 @@ class GrokImageNode(IO.ComfyNode):
resolution=resolution.lower(),
),
response_model=ImageGenerationResponse,
- price_extractor=_extract_grok_price,
)
if len(response.data) == 1:
return IO.NodeOutput(await download_url_to_image_tensor(response.data[0].url))
@@ -351,7 +350,6 @@ class GrokImageEditNode(IO.ComfyNode):
aspect_ratio=None if aspect_ratio == "auto" else aspect_ratio,
),
response_model=ImageGenerationResponse,
- price_extractor=_extract_grok_price,
)
if len(response.data) == 1:
return IO.NodeOutput(await download_url_to_image_tensor(response.data[0].url))
@@ -488,7 +486,6 @@ class GrokImageEditNodeV2(IO.ComfyNode):
aspect_ratio=None if aspect_ratio == "auto" else aspect_ratio,
),
response_model=ImageGenerationResponse,
- price_extractor=_extract_grok_price,
)
if len(response.data) == 1:
return IO.NodeOutput(await download_url_to_image_tensor(response.data[0].url))
diff --git a/comfy_api_nodes/nodes_openai.py b/comfy_api_nodes/nodes_openai.py
index de2c94353..e73319e84 100644
--- a/comfy_api_nodes/nodes_openai.py
+++ b/comfy_api_nodes/nodes_openai.py
@@ -364,19 +364,6 @@ class OpenAIDalle3(IO.ComfyNode):
return IO.NodeOutput(await validate_and_cast_response(response))
-def calculate_tokens_price_image_1(response: OpenAIImageGenerationResponse) -> float | None:
- # https://platform.openai.com/docs/pricing
- return ((response.usage.input_tokens * 10.0) + (response.usage.output_tokens * 40.0)) / 1_000_000.0
-
-
-def calculate_tokens_price_image_1_5(response: OpenAIImageGenerationResponse) -> float | None:
- return ((response.usage.input_tokens * 8.0) + (response.usage.output_tokens * 32.0)) / 1_000_000.0
-
-
-def calculate_tokens_price_image_2_0(response: OpenAIImageGenerationResponse) -> float | None:
- return ((response.usage.input_tokens * 8.0) + (response.usage.output_tokens * 30.0)) / 1_000_000.0
-
-
class OpenAIGPTImage1(IO.ComfyNode):
@classmethod
@@ -570,15 +557,10 @@ class OpenAIGPTImage1(IO.ComfyNode):
if size not in ("auto", "1024x1024", "1024x1536", "1536x1024"):
raise ValueError(f"Resolution {size} is only supported by GPT Image 2 model")
- if model == "gpt-image-1":
- price_extractor = calculate_tokens_price_image_1
- elif model == "gpt-image-1.5":
- price_extractor = calculate_tokens_price_image_1_5
- elif model == "gpt-image-2":
- price_extractor = calculate_tokens_price_image_2_0
+ if model == "gpt-image-2":
if background == "transparent":
raise ValueError("Transparent background is not supported for GPT Image 2 model")
- else:
+ elif model not in ("gpt-image-1", "gpt-image-1.5"):
raise ValueError(f"Unknown model: {model}")
if image is not None:
@@ -633,7 +615,6 @@ class OpenAIGPTImage1(IO.ComfyNode):
),
content_type="multipart/form-data",
files=files,
- price_extractor=price_extractor,
)
else:
response = await sync_op(
@@ -650,7 +631,6 @@ class OpenAIGPTImage1(IO.ComfyNode):
size=size,
moderation="low",
),
- price_extractor=price_extractor,
)
return IO.NodeOutput(await validate_and_cast_response(response))
@@ -879,13 +859,7 @@ class OpenAIGPTImageNodeV2(IO.ComfyNode):
)
size = f"{custom_width}x{custom_height}"
- if model_id == "gpt-image-1":
- price_extractor = calculate_tokens_price_image_1
- elif model_id == "gpt-image-1.5":
- price_extractor = calculate_tokens_price_image_1_5
- elif model_id == "gpt-image-2":
- price_extractor = calculate_tokens_price_image_2_0
- else:
+ if model_id not in ("gpt-image-1", "gpt-image-1.5", "gpt-image-2"):
raise ValueError(f"Unknown model: {model_id}")
if image_tensors:
@@ -944,7 +918,6 @@ class OpenAIGPTImageNodeV2(IO.ComfyNode):
),
content_type="multipart/form-data",
files=files,
- price_extractor=price_extractor,
)
else:
response = await sync_op(
@@ -960,7 +933,6 @@ class OpenAIGPTImageNodeV2(IO.ComfyNode):
size=size,
moderation="low",
),
- price_extractor=price_extractor,
)
return IO.NodeOutput(await validate_and_cast_response(response))
diff --git a/comfy_api_nodes/nodes_openrouter.py b/comfy_api_nodes/nodes_openrouter.py
index e9d6290c2..ee93a1228 100644
--- a/comfy_api_nodes/nodes_openrouter.py
+++ b/comfy_api_nodes/nodes_openrouter.py
@@ -159,12 +159,6 @@ def _build_model_options() -> list[IO.DynamicCombo.Option]:
return [IO.DynamicCombo.Option(spec.slug, _inputs_for_model(spec)) for spec in MODELS]
-def _calculate_price(response: OpenRouterChatResponse) -> float | None:
- if response.usage and response.usage.cost is not None:
- return float(response.usage.cost) * 1.43
- return None
-
-
def _price_badge_jsonata() -> str:
rates_pairs = []
for spec in MODELS:
@@ -372,7 +366,6 @@ class OpenRouterLLMNode(IO.ComfyNode):
ApiEndpoint(path=OPENROUTER_CHAT_ENDPOINT, method="POST"),
response_model=OpenRouterChatResponse,
data=request,
- price_extractor=_calculate_price,
)
return IO.NodeOutput(_extract_text(response))
diff --git a/comfy_api_nodes/nodes_reve.py b/comfy_api_nodes/nodes_reve.py
index 177349a8b..9120c7195 100644
--- a/comfy_api_nodes/nodes_reve.py
+++ b/comfy_api_nodes/nodes_reve.py
@@ -62,13 +62,6 @@ def _postprocessing_inputs():
]
-def _reve_price_extractor(headers: dict) -> float | None:
- credits_used = headers.get("x-reve-credits-used")
- if credits_used is not None:
- return float(credits_used) / 524.48
- return None
-
-
def _reve_response_header_validator(headers: dict) -> None:
error_code = headers.get("x-reve-error-code")
if error_code:
@@ -180,7 +173,6 @@ class ReveImageCreateNode(IO.ComfyNode):
headers={"Accept": "image/webp"},
),
as_binary=True,
- price_extractor=_reve_price_extractor,
response_header_validator=_reve_response_header_validator,
data=ReveImageCreateRequest(
prompt=prompt,
@@ -279,7 +271,6 @@ class ReveImageEditNode(IO.ComfyNode):
headers={"Accept": "image/webp"},
),
as_binary=True,
- price_extractor=_reve_price_extractor,
response_header_validator=_reve_response_header_validator,
data=ReveImageEditRequest(
edit_instruction=edit_instruction,
@@ -396,7 +387,6 @@ class ReveImageRemixNode(IO.ComfyNode):
headers={"Accept": "image/webp"},
),
as_binary=True,
- price_extractor=_reve_price_extractor,
response_header_validator=_reve_response_header_validator,
data=ReveImageRemixRequest(
prompt=prompt,
diff --git a/comfy_api_nodes/util/client.py b/comfy_api_nodes/util/client.py
index 66aab17f8..039e97d58 100644
--- a/comfy_api_nodes/util/client.py
+++ b/comfy_api_nodes/util/client.py
@@ -2,8 +2,10 @@ import asyncio
import contextlib
import json
import logging
+import math
import time
import uuid
+import weakref
from collections.abc import Callable, Iterable
from dataclasses import dataclass
from enum import Enum
@@ -84,11 +86,37 @@ class _PollUIState:
_RETRY_STATUS = {408, 500, 502, 503, 504} # status 429 is handled separately
_MAX_RETRY_AFTER_WAIT = 150.0 # Cap a server Retry-After at this many seconds so a large hint can't block execution
+
+PRICE_CREDITS_HEADER = "X-Comfy-Credits-Used"
+"""Proxy response header with the actual cost in Comfy credits. When present on any successful proxied response,
+it takes precedence over ``price_extractor``."""
+
+_credits_used_by_execution: "weakref.WeakKeyDictionary[type, float]" = weakref.WeakKeyDictionary()
+"""Last PRICE_CREDITS_HEADER value per node execution, keyed by the node's per-execution class clone."""
COMPLETED_STATUSES = ["succeeded", "succeed", "success", "completed", "finished", "done", "complete"]
FAILED_STATUSES = ["cancelled", "canceled", "canceling", "fail", "failed", "error"]
QUEUED_STATUSES = ["created", "queued", "queueing", "submitted", "initializing", "wait", "in_queue"]
+def _maybe_remember_credits_used(node_cls: type[IO.ComfyNode], header_value: str | None) -> None:
+ """Remember a PRICE_CREDITS_HEADER value from a successful proxied response."""
+ if not header_value:
+ return
+ try:
+ credits_used = float(header_value)
+ except (TypeError, ValueError):
+ logging.debug("Ignoring malformed %s header: %r", PRICE_CREDITS_HEADER, header_value)
+ return
+ if not math.isfinite(credits_used) or credits_used < 0:
+ logging.debug("Ignoring out-of-range %s header: %r", PRICE_CREDITS_HEADER, header_value)
+ return
+ _credits_used_by_execution[node_cls] = credits_used + 0.0 # normalize -0.0
+
+
+def _get_remembered_credits_used(node_cls: type[IO.ComfyNode]) -> float | None:
+ return _credits_used_by_execution.get(node_cls)
+
+
async def sync_op(
cls: type[IO.ComfyNode],
endpoint: ApiEndpoint,
@@ -450,10 +478,15 @@ def _display_text(
display_lines: list[str] = []
if status:
display_lines.append(f"Status: {status.capitalize() if isinstance(status, str) else status}")
- if price is not None:
+ server_credits = _get_remembered_credits_used(node_cls)
+ if server_credits is not None:
+ p = f"{server_credits:,.2f}".rstrip("0").rstrip(".")
+ elif price is not None:
p = f"{float(price) * 211:,.1f}".rstrip("0").rstrip(".")
- if p != "0":
- display_lines.append(f"Price: {p} credits")
+ else:
+ p = None
+ if p is not None and p != "0":
+ display_lines.append(f"Price: {p} credits")
if text is not None:
display_lines.append(text)
if display_lines:
@@ -606,7 +639,8 @@ async def _request_base(cfg: _RequestConfig, expect_binary: bool):
"""Core request with retries, per-second interruption monitoring, true cancellation, and friendly errors."""
url = cfg.endpoint.path
parsed_url = urlparse(url)
- if not parsed_url.scheme and not parsed_url.netloc: # is URL relative?
+ is_comfy_api_request = not parsed_url.scheme and not parsed_url.netloc # is URL relative?
+ if is_comfy_api_request:
url = urljoin(default_base_url().rstrip("/") + "/", url.lstrip("/"))
method = cfg.endpoint.method
@@ -644,7 +678,7 @@ async def _request_base(cfg: _RequestConfig, expect_binary: bool):
logging.debug("[DEBUG] HTTP %s %s (attempt %d)", method, url, attempt)
payload_headers = {"Accept": "*/*"} if expect_binary else {"Accept": "application/json"}
- if not parsed_url.scheme and not parsed_url.netloc: # is URL relative?
+ if is_comfy_api_request:
payload_headers.update(get_comfy_api_headers(cfg.node_cls))
if cfg.endpoint.headers:
payload_headers.update(cfg.endpoint.headers)
@@ -804,6 +838,8 @@ async def _request_base(cfg: _RequestConfig, expect_binary: bool):
)
bytes_payload = bytes(buff)
resp_headers = {k.lower(): v for k, v in resp.headers.items()}
+ if is_comfy_api_request:
+ _maybe_remember_credits_used(cfg.node_cls, resp.headers.get(PRICE_CREDITS_HEADER))
if cfg.price_extractor:
with contextlib.suppress(Exception):
extracted_price = cfg.price_extractor(resp_headers)
@@ -831,6 +867,8 @@ async def _request_base(cfg: _RequestConfig, expect_binary: bool):
except json.JSONDecodeError:
payload = {"_raw": text}
response_content_to_log = payload if isinstance(payload, dict) else text
+ if is_comfy_api_request:
+ _maybe_remember_credits_used(cfg.node_cls, resp.headers.get(PRICE_CREDITS_HEADER))
with contextlib.suppress(Exception):
extracted_price = cfg.price_extractor(payload) if cfg.price_extractor else None
operation_succeeded = True