Text Generation
Transformers
Safetensors
English
qwen3
stablecoin
payments
compliance
x402
eip-3009
eip-712
disputes
refunds
agents
agentkit
trl
web3
usdc
fraud-detection
financial-risk
qwen
mcp
ai-agents
prompt-injection
agentic-payments
payment-security
risk-assessment
ofac
conversational
Eval Results (legacy)
Eval Results
text-generation-inference
Instructions to use sriram1983007/SRA-RiskGate-4B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use sriram1983007/SRA-RiskGate-4B with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="sriram1983007/SRA-RiskGate-4B") messages = [ {"role": "user", "content": "Who are you?"}, ] pipe(messages)# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("sriram1983007/SRA-RiskGate-4B") model = AutoModelForCausalLM.from_pretrained("sriram1983007/SRA-RiskGate-4B", device_map="auto") messages = [ {"role": "user", "content": "Who are you?"}, ] inputs = tokenizer.apply_chat_template( messages, add_generation_prompt=True, tokenize=True, return_dict=True, return_tensors="pt", ).to(model.device) outputs = model.generate(**inputs, max_new_tokens=256) print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:])) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use sriram1983007/SRA-RiskGate-4B with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "sriram1983007/SRA-RiskGate-4B" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "sriram1983007/SRA-RiskGate-4B", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/sriram1983007/SRA-RiskGate-4B
- SGLang
How to use sriram1983007/SRA-RiskGate-4B with SGLang:
Install from pip and serve model
# Install SGLang from pip: pip install sglang # Start the SGLang server: python3 -m sglang.launch_server \ --model-path "sriram1983007/SRA-RiskGate-4B" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "sriram1983007/SRA-RiskGate-4B", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker images
docker run --gpus all \ --shm-size 32g \ -p 30000:30000 \ -v ~/.cache/huggingface:/root/.cache/huggingface \ --env "HF_TOKEN=<secret>" \ --ipc=host \ lmsysorg/sglang:latest \ python3 -m sglang.launch_server \ --model-path "sriram1983007/SRA-RiskGate-4B" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "sriram1983007/SRA-RiskGate-4B", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }' - Docker Model Runner
How to use sriram1983007/SRA-RiskGate-4B with Docker Model Runner:
docker model run hf.co/sriram1983007/SRA-RiskGate-4B
Eval results: v2 (2026-11-12)
Browse files
.eval_results/sra-stablecoin-risk-bench.yaml
CHANGED
|
@@ -1,18 +1,18 @@
|
|
| 1 |
- dataset:
|
| 2 |
id: sriram1983007/sra-stablecoin-risk-bench
|
| 3 |
task_id: risk_gate
|
| 4 |
-
value: 0.
|
| 5 |
-
date: '2026-
|
| 6 |
source:
|
| 7 |
-
url: https://huggingface.co/datasets/sriram1983007/sra-bench-results
|
| 8 |
name: Published predictions (self-reported)
|
| 9 |
-
notes: Decision accuracy on 1,175 held-out payment cases; greedy decoding; exact training prompt. Unsafe approval rate 0.
|
| 10 |
- dataset:
|
| 11 |
id: sriram1983007/sra-stablecoin-risk-bench
|
| 12 |
task_id: dispute
|
| 13 |
-
value: 0.
|
| 14 |
-
date: '2026-
|
| 15 |
source:
|
| 16 |
-
url: https://huggingface.co/datasets/sriram1983007/sra-bench-results
|
| 17 |
name: Published predictions (self-reported)
|
| 18 |
-
notes: Outcome accuracy on 825 held-out dispute cases; greedy decoding
|
|
|
|
| 1 |
- dataset:
|
| 2 |
id: sriram1983007/sra-stablecoin-risk-bench
|
| 3 |
task_id: risk_gate
|
| 4 |
+
value: 0.9957
|
| 5 |
+
date: '2026-11-12'
|
| 6 |
source:
|
| 7 |
+
url: https://huggingface.co/datasets/sriram1983007/sra-bench-results/tree/main/sra-riskgate-4b-v2
|
| 8 |
name: Published predictions (self-reported)
|
| 9 |
+
notes: v2. Decision accuracy on 1,175 held-out payment cases; greedy decoding; exact training prompt. Unsafe approval rate 0.0035; over-blocking 0; injected payments approved 2/69. v1 (2026-10-06) scored 0.9949.
|
| 10 |
- dataset:
|
| 11 |
id: sriram1983007/sra-stablecoin-risk-bench
|
| 12 |
task_id: dispute
|
| 13 |
+
value: 0.9018
|
| 14 |
+
date: '2026-11-12'
|
| 15 |
source:
|
| 16 |
+
url: https://huggingface.co/datasets/sriram1983007/sra-bench-results/tree/main/sra-riskgate-4b-v2
|
| 17 |
name: Published predictions (self-reported)
|
| 18 |
+
notes: v2, with the computed dispute window in the prompt (dispute_messages in sra-riskgate 1.3.0+). Outcome accuracy on 825 held-out dispute cases; greedy decoding. Impossible remedy rate 0.0057; wrongful refund rate 0.022; without the window in the prompt 0.8497. v1 (2026-10-06) scored 0.7782.
|