_{BLACKFROST RESEARCH · AUTHORIZED CYBERSECURITY & DEFENSIVE AI}

Kimi-K3.1-Pro-DERISKED

De-risked, cyber-hardened derivative of moonshotai/Kimi-K3.1-Pro · weights pending release · manually gated, unlocked after purchase

Overview · Modification Pipeline · Specifications · Verification Matrix · Gated Access · Discussions

_{CONTEXT CEILING}
1,048,576 ^positions
_{Upstream architectural max (kimi_linear)} _{SPARSE MoE TOPOLOGY}
896 ^{routed experts}
_{16 active / token + 2 shared experts} _{TRANSFORMER DEPTH}
93 ^layers
_{7,168 hidden dim · 96 heads (128 dim)} _{WEIGHT PACKAGE}
1.42 ^TiB
_{96 SafeTensors shards · Omni VLM}

STATUS: Pending weights — no evaluation claims

This repository is a pre-release. The checkpoint has not yet been uploaded, and post-build evaluation is not yet complete. No benchmark, refusal-rate, safety, capability, or throughput number is stated on this card, and none should be inferred from the upstream model's results. When weights and their evaluation are finalized, this card will be updated with the measured results.

Why This Repository Exists

This repository will publish a behaviorally modified Kimi-K3.1-Pro checkpoint intended for authorized security research, red-team evaluation, guardrail testing, and controlled deployment qualification.

The upstream model is a compliance-oriented general assistant. For security work, that orientation is an obstacle: it declines or hedges on the exact benign-but-sensitive behaviors — malware analysis, exploit and payload construction, dual-use tooling, offensive technique discussion — that authorized security professionals need exercised. This derivative is built to remove that obstacle in a controlled, disclosed way.

What Was Done to the Model

Two modifications are applied to the upstream moonshotai/Kimi-K3.1-Pro checkpoint, and nothing else:

_{STAGE 01 · IN-PLACE WEIGHT EDIT}

1. DWM Weight Edits (De-Risking)

An in-place, targeted weight-space edit against the model's refusal direction. The edit subtracts the refusal direction from selected non-expert residual-stream writers while preserving the upstream architecture, expert routing, tokenizer, chat template, and quantization.

Not ablation-by-pruning · Not a retrain · Not a merge

_{STAGE 02 · DOMAIN HARDENING}

2. Blackfrost Cyber SFT

A supervised fine-tune on a curated cybersecurity corpus (dual-use security knowledge, tool reasoning, and authorized-engagement conduct) to harden the model's security-domain competence and operational behavior.

Dual-Use Security · Tool Reasoning · Authorized Conduct

Zero-Sidecar Integrity Guarantee: No runtime patches, serving overlays, system-prompt bakes, or sidecar components ship with this model. The modification is carried entirely in the weights. The private method, recipe, and training data are not distributed.

Interactive Widget · Modification Pipeline & Build Telemetry (2/4 Complete) — click to toggle

Stage	Modification Step	Status	Technical Scope
`STAGE 01 · DWM-SUB`	DWM weight edits (de-risking)	`✓ COMPLETE`	An in-place, targeted weight-space edit against the model's refusal direction on selected non-expert residual-stream writers.
`STAGE 02 · CYBER-SFT`	Blackfrost Cyber SFT	`✓ COMPLETE`	Supervised fine-tune on a curated cybersecurity corpus to harden security-domain competence and operational conduct.
`STAGE 03 · VERIFY`	Post-build evaluation	`⧗ PENDING`	Pending — measured capability, refusal-floor, and long-context evaluation results will be published when complete.
`STAGE 04 · PUBLISH`	Weights upload	`⧗ PENDING`	Pending — 96 SafeTensors shards (~1.42 TiB) staged for gated release upon evaluation sign-off.

Specifications (Inherited from Upstream)

Architecture, tokenizer, chat template, generation configuration, vision tower, and multimodal components are inherited from upstream unmodified in structure. Only the two disclosed weight modifications differ from the upstream checkpoint.

Parameter / Subsystem	Specification
Upstream checkpoint	`moonshotai/Kimi-K3.1-Pro`
Architecture	`KimiK31ForConditionalGeneration (kimi_k31) — hybrid-attention sparse MoE, omni (vision-language)`
Attention mechanism	Hybrid attention architecture (upstream kimi_linear model type)
Layers	`93`
Hidden size	`7,168`
Routed experts	`896 routed · 16 experts per token · 2 shared experts`
Attention heads	`96 (head_dim 128)`
Vocabulary	`163,840 tokens`
Context ceiling	`1,048,576 positions (upstream architectural maximum)`
Modalities	Text and image input → Text output (Omni VLM)
Weight package (reference)	`96 SafeTensors shards · 1,561,217,377,280 indexed bytes (~1.42 TiB)`
Precision	Upstream tensor representation preserved as shipped; see upstream repository for precision details

Cryptographic & Weight Lineage

moonshotai/Kimi-K3.1-Pro  (Upstream Foundation Checkpoint)
  │
  ├── [Stage 01] DWM De-Risking Weight Edits
  │              └─ Refusal-direction subtraction on selected non-expert residual-stream writers
  │              └─ Router gates, 896 MoE experts, vision tower & tokenizer preserved 1:1
  │
  └── [Stage 02] Blackfrost Cyber SFT
                 └─ Cybersecurity-domain supervised fine-tune (dual-use, tool reasoning, conduct)
                 │
                 ▼
       Blackfrost-Research/Kimi-K3.1-Pro-DERISKED   ← THIS REPOSITORY

_{Blackfrost Research is responsible only for this derivative package. Moonshot AI is the upstream model developer and does not endorse this derivative.}

Access: Manually Gated, Unlocked After Purchase

This repository is manually gated, and the checkpoint is unlocked only after purchase. The price is currently unlisted. To obtain access:

_{STEP 01 · COMMERCIAL ARRANGEMENT}
Complete Purchase Agreement
_{Complete the commercial purchase and authorized-scope arrangement directly with Blackfrost Research.}

_{STEP 02 · HUGGING FACE GATE}
Submit Repository Gate Request
_{Submit your access request through the Hugging Face manual gate prompt on this repository page.}

_{STEP 03 · MANUAL UNLOCK}
Verification & Provisioning
_{Access is granted manually after the purchase and organization identity are confirmed on our side.}

License terms for access requests are the upstream Kimi K3.1 Pro License terms (see License). Manual gating and purchase gating are access controls, not safety guarantees.

Verification and Evaluation Status

Qualification Progress: [███░░░░░░░░░░░░░░░░░] 1/6 Checks Complete

Interactive Widget · Pre-Release Verification & Qualification Gate Matrix — click to toggle

Verification Check	Domain	Status	Acceptance & Audit Scope
Repository and card scope audit	`AUDIT`	`✓ COMPLETE`	Lineage disclosure, upstream license attribution, and non-endorsement boundaries verified.
Post-build weight-package validation	`AUDIT`	`⧗ PENDING`	Shard integrity across 96 SafeTensors files (1,561,217,377,280 bytes) and tensor key parity.
General capability evaluation	`CAPABILITY`	`⧗ PENDING`	Reasoning retention, instruction following, code synthesis, and multi-step tool execution.
Safety and behavioral evaluation	`SAFETY`	`⧗ PENDING`	Benign-but-sensitive cybersecurity task completion and calibration profile.
Red-line floor verification	`SAFETY`	`⧗ PENDING`	Confirmation that mandatory Blackfrost release red lines remain intact.
Long-context and vision qualification	`QUALIFICATION`	`⧗ PENDING`	1,048,576 context window stability and multimodal vision-tower alignment verification.

_{No benchmark, throughput, refusal-rate, safety, or capability claim is made for this derivative at this time. Upstream results must not be treated as measured results for this checkpoint.}

Infrastructure & Serving Reference (Post-Unlock)

Interactive Widget · Multi-Node VRAM Sizing & Serving Configurations (SGLang / vLLM) — click to expand

Full Upstream-Precision Reference Cluster

Precision Profile: Native Shipped Precision (96 Shards)
Memory Footprint: ~1.42 TiB indexed weights + KV cache headroom
Target Topology: 2× 8×H200 (141GB) or 2× 8×B200 NVLink Node Cluster
Operational Note: Preserves exact upstream tensor representation across all 93 layers and 896 routed experts.

sglang.launch_server --model-path Blackfrost-Research/Kimi-K3.1-Pro-DERISKED --tp 16 --ep 16 --trust-remote-code --context-length 1048576

High-Density Research Assessment Node

Precision Profile: Multi-Node Tensor + Expert Parallelism (TP8/PP2)
Memory Footprint: ~1,600 GB aggregate cluster memory
Target Topology: 16× H100 80GB SXM5 (InfiniBand RoCEv2)
Operational Note: Recommended for isolated air-gapped security evaluation and Cyber-Frost Harness validation.

vllm serve Blackfrost-Research/Kimi-K3.1-Pro-DERISKED --tensor-parallel-size 8 --pipeline-parallel-size 2 --enable-expert-parallel --max-model-len 262144

Operational Scope: Intended Use vs. Limitations

Intended Use (Authorized Scope)

Authorized Offensive & Purple-Team Research: Authorized offensive-security, red-team, and purple-team research connecting controlled assessment to defensive telemetry.
AI-Safety & Guardrail Stress-Testing: Systematic evaluation of defensive classifiers, blue-team detection pipelines, and guardrail robustness.
Malware Analysis & Exploit Research: Reverse engineering, crash triage, payload mechanics analysis, and dual-use security tooling development in isolated labs.
Controlled Agent & Tool-Use Testing: Evaluating autonomous security harnesses and multi-step CLI/debugger tool invocation under explicit authorization.

Strict Limitations & Operational Risks

This is not a safety-stock model: De-risking reduces refusal behavior, including on benign-but-sensitive content; outputs may include material consumer models decline. Outputs are untrusted and require independent controls, access restrictions, logging, and human review.
Evaluation is pending: Until post-build evaluation is published on this card, treat capability, safety, and stability as unmeasured.
Not a drop-in production release: The model is provided strictly for controlled research and evaluation under the stated manual purchase-gated access conditions.

License and Attribution

This derivative is distributed under the upstream Kimi K3.1 Pro License (Copyright © 2026 Moonshot AI), included as LICENSE when weights publish. That license grants use, copying, modification, distribution, sublicensing, sale, deployment, fine-tuning, and derivative works; this release is a derivative work under it.

See the moonshotai/Kimi-K3.1-Pro repository for upstream documentation and attribution.

The information on this page is provided as-is. For questions or responsible reporting, use this repository's Discussions.