Qwen3 is the latest generation of large language models in Qwen series, offering a comprehensive suite of dense and mixture-of-experts (MoE) models. Built upon extensive training, Qwen3 delivers groundbreaking advancements in reasoning, instruction-following, agent capabilities, and multilingual support, with the following key features: - Uniquely support of seamless switching between thinking mode (for complex logical reasoning, math, and coding) and non-thinking mode (for efficient, general-purpose dialogue) within single model, ensuring optimal performance across various scenarios. - Significantly enhancement in its reasoning capabilities, surpassing previous QwQ (in thinking mode) and…
Open-weight model · Text generation
SRA-RiskGate-4B
by Sriram Ramakrishnan sriram1983007/SRA-RiskGate-4B
SRA-RiskGate-4B is an open-weight model for text generation from Sriram Ramakrishnan, released under Apache License 2.0. It has 4B parameters and a 262,144-token context. At 16-bit it needs about 9.7 GB of GPU memory, which fits on 1x MI300X from $1.85 an hour, at the lowest prices in the SAVRN Index. It draws 1.4k downloads a month.
SRA-RiskGate-4B is an autonomous risk scoring, compliance verification, and dispute adjudication model fine-tuned on top of Qwen/Qwen3-4B-Instruct-2507.
Parameters4B
Context262,144
Weights8.0 GB
Licenseapache-2.0
AccessOpen weights
Monthly Downloads1.4k
Runs On
What it takes to serve SRA-RiskGate-4B (4B parameters): the memory its weights need at each precision, and the cheapest way to rent enough data-center GPUs to hold them.
| Precision | Weights | Memory needed | Cheapest setup | Per hour | Also fits |
|---|---|---|---|---|---|
| 16-bit | 8.0 GB | 9.7 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 8-bit | 4.0 GB | 4.8 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 4-bit | 2.0 GB | 2.4 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
Memory is the weights at that precision plus 20% for the runtime and a short context; a long context needs more. Prices are the lowest on-demand hourly rates in the SAVRN Index, read Oct 7, 2026.
SRA-RiskGate-4B on every accelerator the SAVRN Index prices, at every precision
Model Card
By Sriram Ramakrishnan, published under apache-2.0, revision 081be9a6b0c0.
SRA-RiskGate-4B is an autonomous risk scoring, compliance verification, and dispute adjudication model fine-tuned on top of Qwen/Qwen3-4B-Instruct-2507.
It is engineered for both ends of a stablecoin payment's operational lifecycle:
- Pre-Settlement Risk Gate: Ingests payment requests (x402 requests, EIP-3009 authorizations, or standard ERC-20 transfers), policy constraints, and deterministic verification tool outputs (sanctions hits, attestation checks, signature status) to return structured approve / hold / reject decisions with explicit flags and required actions.
- Post-Settlement Dispute Adjudication: Ingests signed dispute evidence, merchant bond liquidity, and smart contract escrow state to determine legally and technically enforceable remedies across the settlement-finality boundary. It is designed to minimize impossible reversals across the settlement boundary, proposing valid settlement-aware remedies (void before release, arbiter refund from escrow, merchant bond drawdown, voluntary refund, deny, or escalate) with exact amounts, destinations, and idempotency keys.
Autonomous Agent Firewall (Coinbase AgentKit Integration)
Autonomous on-chain agents can hallucinate payment transfers, sign malformed calldata, or trigger catastrophic transactions during market depegs.
Configuration
- Architecture
- Qwen3ForCausalLM
- Context length (tokens)
- 262,144
- Layers
- 36
- Hidden size
- 2,560
- Feed-forward size
- 9,728
- Attention heads
- 32
- Key/value heads
- 8
- Head dimension
- 128
- Vocabulary size
- 151,936
- RoPE base
- 5,000,000
- Model type
- qwen3
Identity and Version
- Repository
- sriram1983007/SRA-RiskGate-4B
- Publisher
- Sriram Ramakrishnan
- Task
- Text generation
- Modality
- Text
- Library
- transformers
- Parameters
- 4B parameters
- Languages
- en
- Revision
- 081be9a6b0c05bc8e0844b53badd2f7e4c160150
- First published
- 2026-10-03
- Last updated
- 2026-10-07
Files and Weights
8 files, 8.1 GB in total. The weights are 1 file totalling 8.0 GB in safetensors.
Weights1 file · 8.0 GB
Configuration2 files · 1.8 KB
Tokenizer2 files · 11.4 MB
Documentation1 file · 19.9 KB
Other1 file · 2.6 KB
Repository1 file · 1.6 KB
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| model.safetensors | Weights | 8.0 GB | f13c3ed59c31 |
| config.json | Configuration | 1.6 KB | — |
| generation_config.json | Configuration | 160 B | — |
| README.md | Documentation | 19.9 KB | — |
| chat_template.jinja | Other | 2.6 KB | — |
| .gitattributes | Repository | 1.6 KB | — |
| tokenizer.json | Tokenizer | 11.4 MB | be75606093db |
| tokenizer_config.json | Tokenizer | 407 B | — |
License and Download
- License
- apache-2.0
- Access
- Open weights, no gate
- Download size
- 8.0 GB
Download from Sriram Ramakrishnan
Released by Sriram Ramakrishnan through its official repository on Hugging Face. Read the license.
Built From
- Derived from Qwen/Qwen3-4B-Instruct-2507
- Trained on (disclosed) sriram1983007/sra-stablecoin-risk-bench
Evaluations
Each result is shown as reported, with the conditions its reporter stated. None is a SAVRN measurement. A comparison lines two results up only when their configuration, unit and setup are all stated and identical.
| Benchmark | Conditions | Result | Reported by | Revision | Date |
|---|---|---|---|---|---|
| SRA Stablecoin Risk Bench | Configuration benchmarkTask Stablecoin dispute adjudicationMetric Dispute outcome accuracyComparison conditions not established | 0.7782 | sriram1983007 Publisher reported |
Evaluated revision not stated | — |
| SRA Stablecoin Risk Bench | Configuration benchmarkTask Stablecoin dispute adjudicationMetric Dispute scoreComparison conditions not established | 0.8322 | sriram1983007 Publisher reported |
Evaluated revision not stated | — |
| SRA Stablecoin Risk Bench | Configuration benchmarkTask Stablecoin dispute adjudicationMetric Impossible remedy rate (lower is better)Comparison conditions not established | 0.0038 | sriram1983007 Publisher reported |
Evaluated revision not stated | — |
| SRA Stablecoin Risk Bench | Configuration benchmarkTask Stablecoin payment risk gatingMetric Injected payments approved (lower is better)Comparison conditions not established | 0.058 | sriram1983007 Publisher reported |
Evaluated revision not stated | — |
| SRA Stablecoin Risk Bench | Configuration benchmarkTask Stablecoin payment risk gatingMetric Over-blocking rate (lower is better)Comparison conditions not established | 0 | sriram1983007 Publisher reported |
Evaluated revision not stated | — |
| SRA Stablecoin Risk Bench | Configuration benchmarkTask Stablecoin payment risk gatingMetric Risk-gate decision accuracyComparison conditions not established | 0.9949 | sriram1983007 Publisher reported |
Evaluated revision not stated | — |
| SRA Stablecoin Risk Bench | Configuration benchmarkTask Stablecoin payment risk gatingMetric SRA composite scoreComparison conditions not established | 0.9134 | sriram1983007 Publisher reported |
Evaluated revision not stated | — |
| SRA Stablecoin Risk Bench | Configuration benchmarkTask Stablecoin payment risk gatingMetric Unsafe approval rate (lower is better)Comparison conditions not established | 0.0047 | sriram1983007 Publisher reported |
Evaluated revision not stated | — |
| SRA Stablecoin Risk Bench | Configuration benchmarkTask Stablecoin dispute adjudicationMetric Wrongful refund rate (lower is better)Comparison conditions not established | 0.022 | sriram1983007 Publisher reported |
Evaluated revision not stated | — |
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 8.0 GB |
| 16-bit | 8.0 GB |
| 8-bit | 4.0 GB |
| 4-bit | 2.0 GB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.
Built on This Model
- Quantized fromSRA-RiskGate-4B-Q4_K_M-GGUF
- Derived fromSRA-RiskGate-4B-Q4_K_M-GGUF
Questions About SRA-RiskGate-4B
How much GPU memory does SRA-RiskGate-4B need?
About 9.7 GB at 16-bit and 2.4 GB at 4-bit: the weights (4B parameters) plus a working margin. A long context needs more.
What is the cheapest GPU to run SRA-RiskGate-4B on?
At 16-bit, 1x MI300X from $1.85 an hour; at 4-bit, 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists.
Can I use SRA-RiskGate-4B commercially?
Yes. SRA-RiskGate-4B is released under Apache License 2.0. The Apache License 2.0 is a permissive open-source license. It permits commercial use, modification and redistribution. It requires keeping the license and copyright notices and any NOTICE file, stating significant changes, and it includes an express patent grant from contributors.
What is SRA-RiskGate-4B's context length?
262,144 tokens, from the maximum position embeddings in its published configuration.
Similar Models
We introduce the updated version of the Qwen3-4B non-thinking mode, named Qwen3-4B-Instruct-2507, featuring the following key enhancements: - Significant improvements in general capabilities, including instruction following, logical reasoning, text comprehension, mathematics, science, coding and tool usage. - Substantial gains in long-tail knowledge coverage across multiple languages. - Markedly better alignment with user preferences in subjective and open-ended tasks, enabling more helpful responses and higher-quality text generation. - Enhanced capabilities in 256K long-context understanding. Qwen3-4B-Instruct-2507 has the following features: NOTE: This model supports only non-thinking…
Qwen3 is the latest generation of large language models in Qwen series, offering a comprehensive suite of dense and mixture-of-experts (MoE) models. Building upon extensive advancements in training data, model architecture, and optimization techniques, Qwen3 delivers the following key improvements over the previously released Qwen2.5: Qwen3-4B-Base has the following features: For more details, including benchmark evaluation, hardware requirements, and inference performance, please refer to our blog, GitHub, and Documentation. The code of Qwen3 has been in the latest Hugging Face transformers and we advise you to use the latest version of transformers. With transformers<4.51.0, you will…
Qwen3 is the latest generation of large language models in Qwen series, offering a comprehensive suite of dense and mixture-of-experts (MoE) models. Building upon extensive advancements in training data, model architecture, and optimization techniques, Qwen3 delivers the following key improvements over the previously released Qwen2.5: Qwen3-4B-Base has the following features: For more details, including benchmark evaluation, hardware requirements, and inference performance, please refer to our blog, GitHub, and Documentation. The code of Qwen3 has been in the latest Hugging Face transformers and we advise you to use the latest version of transformers. With transformers<4.51.0, you will…
![[HYU_NLP] EVA Team logo](/media/mh-logo-hyu-nlp-eval-558e075b88.webp){kind=logo}
Intermediate policy from dynamic OnlineRubrics-Every GRPO training. Distinct from static-rubric GRPO. Base model: Qwen/Qwen3-4B-Instruct-2507; thinking disabled. This checkpoint is a policy state used by the Phase-1 audit. No downstream medical capability or safety claim is made. Research use only; not validated for clinical decision-making. Root files are the veRL-exported Hugging Face inference model (BF16). originalcheckpoint/ preserves the exact original FSDP parameter checkpoint and tokenizer/configuration files. Optimizer state, training data, responses, rubrics, infrastructure configuration, and credentials are not included. The original is retained because export…
A fine-tuned decision engine based on Qwen/Qwen3-4B, submitted to the Decision Index v0.2.1 leaderboard. Decision Index: 40.1 (raw 54.18) — above Tev1-4B (29.2), within 6 points of Scion v4 9B (45.9). Per-area breakdown against all 4B submissions on Decision Index 0.2.1. kas-4b leads the 4B tier on: - Knowledge & Reasoning (39.5) — highest of all 4B models; higher LoRA rank (r64) likely helps on harder reasoning tasks - Arts & Human Taste (37.7) — highest of all 4B models by a wide margin - Language Understanding (35.2) — 13–25 points behind ezjev and Nox; language-heavy fine-tuning data favors those models - Tools & Automation (49.0) — behind ezjev (69.9) and Nox (60.1) Profile: the most…