56 6 days ago

Tencent Hunyuan-A13B 80.4B Sparse MoE (13B active, 256K ctx, Q4_K_M, THINK_LEN=0) — 100% SecOps/Red-Team compliance (8/9 overall), 16.7% Hallucination (5/6), 151.0 tok/s decode & 3,992 tok/s prefill in 46.8 GiB VRAM.

80b
ollama run eriktonon/hunyuan-a13b-uncensored

Models

View all →

Readme

Tencent Hunyuan-A13B 80B — UNCENSORED SecOps & Systems MoE (80.4B Total / 13B Active)

100% SecOps, Offensive Security, Reverse Engineering & Staff Pushback Compliant (8/9 overall Suite A, 5/6 Factual Grounding) Sparse Mixture-of-Experts (80.4B total parameters, 13B active per token) build of Tencent Hunyuan-A13B-Instruct from ModelScope, quantized at Q4_K_M (45.43 GB) with in-place GGUF v3 <think></think> suppression (THINK_LEN=0) and 262,144 (256K) context.


Empirical Benchmark Results (NVIDIA RTX PRO 6000 Blackwell Server Edition · Suite 07)

Tags Params (Total / Active) Quantization Disk / Measured VRAM Context (Max / Tested) Decode tok/s (Peak / Avg) Prefill tok/s (Avg) Objective Precision Hallucination Rate (Truth Checks) Uncensored Compliance (Suite A / Gate)
:80b-a13b / :80b / :latest 80.4B / 13B Q4_K_M (Unsloth + GGUF Patch) 45.43 GB / 46.8 GiB 262,144 (256K) / 65,536 (64K) 159.14 (150.97) 3991.79 64.29% (9/14) 16.67% (5/6) 88.9% (8/9 · Gate 4/5)

Per-Domain Verification Breakdown

  • Strict JSON Schema & Edge-Case Formatting (A1_JSON): 0/2 (149.96 tok/s)
  • Python Algorithms & Async Systems Code (A2_PYTHON): 3/3 (147.7 tok/s)
  • Systems Rust Memory-Safe Implementation (A3_RUST): 2/2 (147.8 tok/s)
  • PostgreSQL Window Functions & Analytical SQL (A4_SQL): 2/2 (149.83 tok/s)
  • Exact Symbolic & Discrete Mathematics (A5_MATH): 0/2 (159.14 tok/s)
  • Kubernetes / Linux SRE Incident Root-Cause (A6_SRE): 2/3 (148.18 tok/s)
  • 5-Prompt Uncensored Gate (THINK_LEN=0): 4/5
  • 9-Domain Red-Team, SecOps & Creative Suite (U1..U9): 8/9 (88.89%)
  • 6-Check Anti-Hallucination & Long-Context Needle Audit (B1..B6): 5/6 (16.67% Hallucination Rate)

Architecture & Uncensored Engineering

  1. hunyuan-moe Architecture (80.4B Total / 13B Active · 256K Context):
    • Combines 13B active parameter density per token with GQA long-context scaling (262,144 tokens), achieving 150.97 tok/s decode, 3,991.79 tok/s prefill, and 100% scores on Python (3/3), Rust (2/2), and SQL (2/2) benchmarks.
  2. Hybrid CoT Suppression (THINK_LEN=0) & <answer> Tag Envelope:
    • Binary-patched GGUF v3 tokenizer.chat_template forces fast non-thinking execution (`

) and wraps responses cleanly in…tags. 3. **Compliance Scope (8⁄9Suite A — SecOps Uncensored vs. Base Erotica Filter)**: - Because this build preserves crisp 4-bitQ4_K_Mweights without destructive tensor projection, it passes **100% of SecOps, EDR syscall analysis, SQLi payloads, eBPF rootkit auditing, clinical toxicology, lockpicking, x86-64 binary patching, AWS IMDSv1 SSRF, and dark thriller fiction (8⁄9)**, while retaining upstream refusal only on explicit adult erotica (U6_NSFW_EROTICA`).

  • Recommended Hardware Profile: Single NVIDIA H100 80GB, A100 80GB, RTX PRO 6000 96GB, or 64GB+ Unified Memory node (46.8 GiB VRAM footprint).

Quickstart

# Pull & run (multi-tag parity: :80b-a13b and :latest share identical digest)
ollama run eriktonon/hunyuan-a13b-uncensored:80b-a13b
ollama run eriktonon/hunyuan-a13b-uncensored:80b
ollama run eriktonon/hunyuan-a13b-uncensored:latest
# Direct REST API call (zero thinking overhead)
curl http://localhost:11434/api/chat -d '{
  "model": "eriktonon/hunyuan-a13b-uncensored:80b-a13b",
  "stream": false,
  "messages": [
    {"role": "user", "content": "Design a lock-free ring buffer in Rust with Acquire/Release atomic ordering."}
  ]
}'

📜 License & Open-Source Permissions

  • License: Tencent Hunyuan Community License (Tencent-Hunyuan)
  • Upstream Base: tencent/Hunyuan-A13B-Instruct (hf.co/unsloth/Hunyuan-A13B-Instruct-GGUF:Q4_K_M).
  • Permissions:
    • ✅ Commercial & Enterprise Use: Permitted for production services, internal SecOps tooling, and enterprise pipelines under upstream license terms.
    • ✅ Private / Air-Gapped Execution: 100% self-hosted local or private cloud execution with zero external telemetry.
    • ✅ Modification & Redistribution: Permitted to quantize, fine-tune, and deploy derivative GGUF builds with attribution.