opengradient 0.5.1__py3-none-any.whl → 0.5.3__py3-none-any.whl

opengradient/cli.py

@@ -20,7 +20,7 @@ from .defaults import (
     DEFAULT_API_URL,
     DEFAULT_LLM_SERVER_URL,
 )
-from .types import InferenceMode, LlmInferenceMode, LLM, TEE_LLM
+from .types import InferenceMode, LlmInferenceMode, LLM, TEE_LLM, x402SettlementMode
 
 OG_CONFIG_FILE = Path.home() / ".opengradient_config.json"
 
@@ -74,6 +74,12 @@ LlmInferenceModes = {
 }
 
 
+x402SettlementModes = {
+    "settle-batch": x402SettlementMode.SETTLE_BATCH,
+    "settle": x402SettlementMode.SETTLE,
+    "settle-metadata": x402SettlementMode.SETTLE_METADATA,
+}
+
 def initialize_config(ctx):
     """Interactively initialize OpenGradient config"""
     if ctx.obj:  # Check if config data already exists
@@ -422,8 +428,9 @@ def infer(ctx, model_cid: str, inference_mode: str, input_data, input_file: Path
 @click.option("--stop-sequence", multiple=True, help="Stop sequences for LLM")
 @click.option("--temperature", type=float, default=0.0, help="Temperature for LLM inference (0.0 to 1.0)")
 @click.option("--local", is_flag=True, help="Force use of local model even if not in LLM enum")
+@click.option("--x402-settlement-mode", "x402_settlement_mode", type=click.Choice(x402SettlementModes.keys()), default="settle-batch", help="Settlement mode for x402 payload")
 @click.pass_context
-def completion(ctx, model_cid: str, inference_mode: str, prompt: str, max_tokens: int, stop_sequence: List[str], temperature: float, local: bool):
+def completion(ctx, model_cid: str, inference_mode: str, x402_settlement_mode: str, prompt: str, max_tokens: int, stop_sequence: List[str], temperature: float, local: bool):
     """
     Run completion inference on an LLM model (local or external).
 
@@ -464,6 +471,7 @@ def completion(ctx, model_cid: str, inference_mode: str, prompt: str, max_tokens
             stop_sequence=list(stop_sequence),
             temperature=temperature,
             local_model=local,
+            x402_settlement_mode=x402_settlement_mode,
         )
 
         print_llm_completion_result(model_cid, completion_output.transaction_hash, completion_output.completion_output, is_local)
@@ -529,6 +537,7 @@ def print_llm_completion_result(model_cid, tx_hash, llm_output, is_local=True):
 )
 @click.option("--tool-choice", type=str, default="", help="Specific tool choice for the LLM")
 @click.option("--local", is_flag=True, help="Force use of local model even if not in LLM enum")
+@click.option("--x402-settlement-mode", type=click.Choice(x402SettlementModes.keys()), default="settle-batch", help="Settlement mode for x402 payload")
 @click.pass_context
 def chat(
     ctx,
@@ -542,6 +551,7 @@ def chat(
     tools: Optional[str],
     tools_file: Optional[Path],
     tool_choice: Optional[str],
+    x402_settlement_mode: Optional[str],
     local: bool,
 ):
     """
@@ -637,6 +647,7 @@ def chat(
             tools=parsed_tools,
             tool_choice=tool_choice,
             local_model=local,
+            x402_settlement_mode=x402_settlement_mode,
         )
 
         print_llm_chat_result(

opengradient/client.py

@@ -23,6 +23,7 @@ from .proto import infer_pb2, infer_pb2_grpc
 from .types import (
     LLM,
     TEE_LLM,
+    x402SettlementMode,
     HistoricalInputQuery,
     InferenceMode,
     LlmInferenceMode,
@@ -436,6 +437,7 @@ class Client:
         inference_mode: LlmInferenceMode = LlmInferenceMode.VANILLA,
         max_retries: Optional[int] = None,
         local_model: Optional[bool] = False,
+        x402_settlement_mode: Optional[x402SettlementMode] = x402SettlementMode.SETTLE_BATCH,
     ) -> TextGenerationOutput:
         """
         Perform inference on an LLM model using completions.
@@ -465,11 +467,12 @@ class Client:
                 return OpenGradientError("That model CID is not supported yet for TEE inference")
 
             return self._external_llm_completion(
-                model=model_cid,
+                model=model_cid.split('/')[1],
                 prompt=prompt,
                 max_tokens=max_tokens,
                 stop_sequence=stop_sequence,
                 temperature=temperature,
+                x402_settlement_mode=x402_settlement_mode,
             )
         
         # Original local model logic
@@ -479,12 +482,16 @@ class Client:
 
             if model_cid not in [llm.value for llm in LLM]:
                 raise OpenGradientError("That model CID is not yet supported for inference")
+            
+            model_name = model_cid
+            if model_cid in [llm.value for llm in TEE_LLM]:
+                model_name = model_cid.split('/')[1]
 
             contract = self._blockchain.eth.contract(address=self._inference_hub_contract_address, abi=self._inference_abi)
 
             llm_request = {
                 "mode": inference_mode.value,
-                "modelCID": model_cid,
+                "modelCID": model_name,
                 "prompt": prompt,
                 "max_tokens": max_tokens,
                 "stop_sequence": stop_sequence or [],
@@ -512,6 +519,7 @@ class Client:
         max_tokens: int = 100,
         stop_sequence: Optional[List[str]] = None,
         temperature: float = 0.0,
+        x402_settlement_mode: Optional[x402SettlementMode] = x402SettlementMode.SETTLE_BATCH,
     ) -> TextGenerationOutput:
         """
         Route completion request to external LLM server with x402 payments.
@@ -580,7 +588,8 @@ class Client:
             ) as client:
                 headers = {
                     "Content-Type": "application/json",
-                    "Authorization": f"Bearer {X402_PLACEHOLDER_API_KEY}"
+                    "Authorization": f"Bearer {X402_PLACEHOLDER_API_KEY}",
+                    "X-SETTLEMENT-TYPE": x402_settlement_mode,
                 }
                 
                 payload = {
@@ -641,6 +650,7 @@ class Client:
         tool_choice: Optional[str] = None,
         max_retries: Optional[int] = None,
         local_model: Optional[bool] = False,
+        x402_settlement_mode: Optional[x402SettlementMode] = x402SettlementMode.SETTLE_BATCH,
     ) -> TextGenerationOutput:
         """
         Perform inference on an LLM model using chat.
@@ -670,13 +680,14 @@ class Client:
                 return OpenGradientError("That model CID is not supported yet for TEE inference")
 
             return self._external_llm_chat(
-                model=model_cid,
+                model=model_cid.split('/')[1],
                 messages=messages,
                 max_tokens=max_tokens,
                 stop_sequence=stop_sequence,
                 temperature=temperature,
                 tools=tools,
                 tool_choice=tool_choice,
+                x402_settlement_mode=x402_settlement_mode,
             )
         
         # Original local model logic
@@ -686,6 +697,10 @@ class Client:
             
             if model_cid not in [llm.value for llm in LLM]:
                 raise OpenGradientError("That model CID is not yet supported for inference")
+            
+            model_name = model_cid
+            if model_cid in [llm.value for llm in TEE_LLM]:
+                model_name = model_cid.split('/')[1]
 
             contract = self._blockchain.eth.contract(address=self._inference_hub_contract_address, abi=self._inference_abi)
 
@@ -713,7 +728,7 @@ class Client:
 
             llm_request = {
                 "mode": inference_mode.value,
-                "modelCID": model_cid,
+                "modelCID": model_name,
                 "messages": messages,
                 "max_tokens": max_tokens,
                 "stop_sequence": stop_sequence or [],
@@ -752,6 +767,7 @@ class Client:
         temperature: float = 0.0,
         tools: Optional[List[Dict]] = None,
         tool_choice: Optional[str] = None,
+        x402_settlement_mode: x402SettlementMode = x402SettlementMode.SETTLE_BATCH,
     ) -> TextGenerationOutput:
         """
         Route chat request to external LLM server with x402 payments.
@@ -827,7 +843,8 @@ class Client:
             ) as client:
                 headers = {
                     "Content-Type": "application/json",
-                    "Authorization": f"Bearer {X402_PLACEHOLDER_API_KEY}"
+                    "Authorization": f"Bearer {X402_PLACEHOLDER_API_KEY}",
+                    "X-SETTLEMENT-TYPE": x402_settlement_mode
                 }
                 
                 payload = {

opengradient/types.py

@@ -1,10 +1,15 @@
 import time
 from dataclasses import dataclass
-from enum import Enum, IntEnum
+from enum import Enum, IntEnum, StrEnum
 from typing import Dict, List, Optional, Tuple, Union, DefaultDict
 import numpy as np
 
 
+class x402SettlementMode(StrEnum):
+    SETTLE = "settle"
+    SETTLE_METADATA = "settle-metadata"
+    SETTLE_BATCH = "settle-batch"
+
 class CandleOrder(IntEnum):
     ASCENDING = 0
     DESCENDING = 1
@@ -202,27 +207,27 @@ class LLM(str, Enum):
     DOBBY_LEASHED_3_1_8B = "SentientAGI/Dobby-Mini-Leashed-Llama-3.1-8B"
     
     # OpenAI models via TEE
-    GPT_4_1_2025_04_14 = "gpt-4.1-2025-04-14"
-    GPT_4O = "gpt-4o"
-    O4_MINI = "o4-mini"
+    GPT_4_1_2025_04_14 = "OpenAI/gpt-4.1-2025-04-14"
+    GPT_4O = "OpenAI/gpt-4o"
+    O4_MINI = "OpenAI/o4-mini"
     
     # Anthropic models via TEE
-    CLAUDE_3_7_SONNET = "claude-3.7-sonnet"
-    CLAUDE_3_5_HAIKU = "claude-3.5-haiku"
-    CLAUDE_4_0_SONNET = "claude-4.0-sonnet"
+    CLAUDE_3_7_SONNET = "Anthropic/claude-3.7-sonnet"
+    CLAUDE_3_5_HAIKU = "Anthropic/claude-3.5-haiku"
+    CLAUDE_4_0_SONNET = "Anthropic/claude-4.0-sonnet"
     
     # Google models via TEE
-    GEMINI_2_5_FLASH = "gemini-2.5-flash"
-    GEMINI_2_5_PRO = "gemini-2.5-pro"
-    GEMINI_2_0_FLASH = "gemini-2.0-flash"
+    GEMINI_2_5_FLASH = "Google/gemini-2.5-flash"
+    GEMINI_2_5_PRO = "Google/gemini-2.5-pro"
+    GEMINI_2_0_FLASH = "Google/gemini-2.0-flash"
     
     # xAI Grok models via TEE
-    GROK_3_MINI_BETA = "grok-3-mini-beta"
-    GROK_3_BETA = "grok-3-beta"
-    GROK_2_1212 = "grok-2-1212"
-    GROK_2_VISION_LATEST = "grok-2-vision-latest"
-    GROK_4_1_FAST = "grok-4.1-fast"
-    GROK_4_1_FAST_NON_REASONING = "grok-4-1-fast-non-reasoning"
+    GROK_3_MINI_BETA = "xAI/grok-3-mini-beta"
+    GROK_3_BETA = "xAI/grok-3-beta"
+    GROK_2_1212 = "xAI/grok-2-1212"
+    GROK_2_VISION_LATEST = "xAI/grok-2-vision-latest"
+    GROK_4_1_FAST = "xAI/grok-4.1-fast"
+    GROK_4_1_FAST_NON_REASONING = "xAI/grok-4-1-fast-non-reasoning"
 
 class TEE_LLM(str, Enum):
     """Enum for LLM models available for TEE execution"""
@@ -231,27 +236,27 @@ class TEE_LLM(str, Enum):
     # META_LLAMA_3_1_70B_INSTRUCT = "meta-llama/Llama-3.1-70B-Instruct"
     
     # OpenAI models via TEE
-    GPT_4_1_2025_04_14 = "gpt-4.1-2025-04-14"
-    GPT_4O = "gpt-4o"
-    O4_MINI = "o4-mini"
+    GPT_4_1_2025_04_14 = "OpenAI/gpt-4.1-2025-04-14"
+    GPT_4O = "OpenAI/gpt-4o"
+    O4_MINI = "OpenAI/o4-mini"
     
     # Anthropic models via TEE
-    CLAUDE_3_7_SONNET = "claude-3.7-sonnet"
-    CLAUDE_3_5_HAIKU = "claude-3.5-haiku"
-    CLAUDE_4_0_SONNET = "claude-4.0-sonnet"
+    CLAUDE_3_7_SONNET = "Anthropic/claude-3.7-sonnet"
+    CLAUDE_3_5_HAIKU = "Anthropic/claude-3.5-haiku"
+    CLAUDE_4_0_SONNET = "Anthropic/claude-4.0-sonnet"
     
     # Google models via TEE
-    GEMINI_2_5_FLASH = "gemini-2.5-flash"
-    GEMINI_2_5_PRO = "gemini-2.5-pro"
-    GEMINI_2_0_FLASH = "gemini-2.0-flash"
+    GEMINI_2_5_FLASH = "Google/gemini-2.5-flash"
+    GEMINI_2_5_PRO = "Google/gemini-2.5-pro"
+    GEMINI_2_0_FLASH = "Google/gemini-2.0-flash"
     
     # xAI Grok models via TEE
-    GROK_3_MINI_BETA = "grok-3-mini-beta"
-    GROK_3_BETA = "grok-3-beta"
-    GROK_2_1212 = "grok-2-1212"
-    GROK_2_VISION_LATEST = "grok-2-vision-latest"
-    GROK_4_1_FAST = "grok-4.1-fast"
-    GROK_4_1_FAST_NON_REASONING = "grok-4-1-fast-non-reasoning"
+    GROK_3_MINI_BETA = "xAI/grok-3-mini-beta"
+    GROK_3_BETA = "xAI/grok-3-beta"
+    GROK_2_1212 = "xAI/grok-2-1212"
+    GROK_2_VISION_LATEST = "xAI/grok-2-vision-latest"
+    GROK_4_1_FAST = "xAI/grok-4.1-fast"
+    GROK_4_1_FAST_NON_REASONING = "xAI/grok-4-1-fast-non-reasoning"
 
 
 @dataclass

{opengradient-0.5.1.dist-info → opengradient-0.5.3.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: opengradient
-Version: 0.5.1
+Version: 0.5.3
 Summary: Python SDK for OpenGradient decentralized model management & inference services
 Author-email: OpenGradient <kyle@vannalabs.ai>
 License-Expression: MIT

{opengradient-0.5.1.dist-info → opengradient-0.5.3.dist-info}/RECORD

@@ -1,10 +1,10 @@
 opengradient/__init__.py,sha256=wVg0KTFNBl7RnZF9huR5-m_q1E7tO-YyQwY7AD9JFoc,12635
 opengradient/account.py,sha256=5wrYpws_1lozjOFjLCTHtxgoxK-LmObDAaVy9eDcJY4,1145
-opengradient/cli.py,sha256=QzjH_KS6TF8gm_L1otFWA-oHkJ5SSfizFoRn0xR0b70,29162
-opengradient/client.py,sha256=46pj3l-JkQfdojoLh94na0kJIYIVwusTcDP5qi_nKwk,62004
+opengradient/cli.py,sha256=RksBEGVcZgUg6ng53Fgz-Ncv1erBwdADgblB2HmKkwk,29868
+opengradient/client.py,sha256=4_Rr5oQ-wrj-LTA54uWF_FplkpkRWT1K2ectD8SSEoA,62991
 opengradient/defaults.py,sha256=w8-dr5ciF2TGnqbm_ib0Yz4U0YL5ikpNqkcPVpmXzP8,673
 opengradient/exceptions.py,sha256=88tfegboGtlehQcwhxsl6ZzhLJWZWlkf_bkHTiCtXpo,3391
-opengradient/types.py,sha256=f4HiDnsNBHtsyLD6AlWihtSB93qTFmj72zOQFZRNK5k,7360
+opengradient/types.py,sha256=0JqnLiUXmqpunpnK3V0vrQz6ymFO1xG9TgAXjyMyxUE,7693
 opengradient/utils.py,sha256=ZUq4OBIml2vsC0tRqus4Zwb_e3g4woo00apByrafuVw,8058
 opengradient/abi/InferencePrecompile.abi,sha256=reepTHg6Q01UrFP0Gexc-JayplsvOLPfG7jrEZ-cV28,10197
 opengradient/abi/PriceHistoryInference.abi,sha256=ZB3fZdx1kaFlp2wt1vTbTZZG1k8HPvmNtkG5Q8Bnajw,5098
@@ -27,9 +27,9 @@ opengradient/workflow_models/constants.py,sha256=viIkb_LGcfVprqQNaA80gBTj6cfYam0
 opengradient/workflow_models/types.py,sha256=Z22hF6c8Y4D2GlzVEIBODGwsqSjSrQvUcpZ7R-mIJdI,409
 opengradient/workflow_models/utils.py,sha256=ySfpuiOBqLTlfto6ZxZf2vc7K6RGIja0l4eaVm5AOzY,1503
 opengradient/workflow_models/workflow_models.py,sha256=d4C_gs39DAfy4cdY9Ee6GMXpPfzwvKFpmxzK1A7LNgU,3900
-opengradient-0.5.1.dist-info/licenses/LICENSE,sha256=xEcvQ3AxZOtDkrqkys2Mm6Y9diEnaSeQRKvxi-JGnNA,1069
-opengradient-0.5.1.dist-info/METADATA,sha256=_Gi-TVU_xljBPwwS6J3olNQWul73Hy3Ak0tErnP1Ed0,3992
-opengradient-0.5.1.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-opengradient-0.5.1.dist-info/entry_points.txt,sha256=yUKTaJx8RXnybkob0J62wVBiCp_1agVbgw9uzsmaeJc,54
-opengradient-0.5.1.dist-info/top_level.txt,sha256=oC1zimVLa2Yi1LQz8c7x-0IQm92milb5ax8gHBHwDqU,13
-opengradient-0.5.1.dist-info/RECORD,,
+opengradient-0.5.3.dist-info/licenses/LICENSE,sha256=xEcvQ3AxZOtDkrqkys2Mm6Y9diEnaSeQRKvxi-JGnNA,1069
+opengradient-0.5.3.dist-info/METADATA,sha256=cIUG2yYqEVmN27jL0gHUaRYmMwg89ZfNiP0gtPg32kg,3992
+opengradient-0.5.3.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+opengradient-0.5.3.dist-info/entry_points.txt,sha256=yUKTaJx8RXnybkob0J62wVBiCp_1agVbgw9uzsmaeJc,54
+opengradient-0.5.3.dist-info/top_level.txt,sha256=oC1zimVLa2Yi1LQz8c7x-0IQm92milb5ax8gHBHwDqU,13
+opengradient-0.5.3.dist-info/RECORD,,