Home

Field SOP

Scenario-based tutorials with real screenshots, prompts, workflow templates and pitfalls.

3 RMB a Brand Video? LibTV WorkBuddy Video Factory SOP

LibTV x WorkBuddy officially connected on October 26: a two-step setup in WorkBuddy (search LibTV in connectors, authorize, then upload materials and state your need), with the need spelled out in four parts — what to shoot, where it runs, what style, and what must be preserved. Three copy-ready scenario templates: brand videos (product docs to selling points to storyboard to finished cut), UGC variants (one proven idea batch-produced with different openings and scenes for ad testing), and episodic short drama (synopsis to plot breakdown to storyboard to continuous shots); paired with scheduled tasks for a 7x24 production line. Costs are official self-reported figures: 3 RMB per brand video, 5 RMB per drama episode (riding the cost edge of Wan 3 and Seedance 2.0 Fast); hit works like Come-Over are published with canvases and prompts for replication; the advanced route runs through the open-source skills repo libtv-labs/libtv-skills (MIT).

Claude Code Sandboxing: Install It Without Giving Agents Root

Give your agent a sandbox, not root: a hands-on install guide for Anthropic's open-source sandboxing tool srt (anthropics/sandbox-runtime, Apache-2.0, 6,615 stars, repo created October 2025). Verbatim npm global install, the ~/.srt-settings.json config verbatim, platform primitive differences across macOS/Linux/Windows with three default-policy tiers, and the three-layer Windows truth stated as-is: the srt CLI is still Alpha, Schannel is a hard limitation with no workaround, and Claude Code on native Windows does not run the sandbox — use WSL2. The 84 percent interception rate is the official self-reported figure; an eight-item community pitfall list included (#139/#160/#122 fixed, Ubuntu 24.04 sysctl, and more).

A 3B Model That Chews 200k Pages a Day: DeepSeek-OCR Deploy SOP

A hands-on DeepSeek-OCR deployment SOP: the vision side squeezes a whole page into a few dozen to a few hundred vision tokens while the 3B MoE decoder keeps only about 570M active parameters — the official paper cites 200k+ pages of training data per day on one A100-40G, and a 24GB consumer card gets you in. The two official repos take different routes: environment and Transformers inference on generation 2 (Apache-2.0), upstream vLLM and API serving on generation 1's mature ecosystem (MIT). Verbatim commands for the cuda11.8+torch2.6.0+flash-attn 2.7.3 stack, a five-mode resolution token ledger (64 to 400+), the NGram anti-repetition processor with the 128821/128822 table whitelist, community-only VRAM guidance (nothing official), and a ten-item troubleshooting checklist.

428B Weights Are Out, Budget 8 H100s: M3.1 Flash Local Deploy

A deployment log for M3.1 Flash (428B, 23B active). Cold water first: there is no consumer or single-GPU recipe anywhere official - the smallest validated tiers are 4x Blackwell (B200/B300/GB200/GB300, TP4), 4x AMD MI355X (TP4), or a full node of 8x H100/H200 (TP8, FP8 path). Weights come from HF or ModelScope (public, not gated); three size calibers coexist: community-count 252.4GB over 48 shards plus a 2.3GB DSpark head, and the vLLM recipe listing NVFP4 at 285GB, MXFP4 at 293GB, FP8 at 533GB; there is no official quantized repo. vLLM needs 0.33.0 or newer (absent from PyPI on release day, so use the vllm/vllm-openai:minimax-m31-flash Docker image or nightly); three official commands transcribed verbatim - H100/H200 TP8, GB200 TP4, MI355X TP4 - all forcing --block-size 128 and --disable-hybrid-kv-cache-manager (omit it and you get No common block size for 16); 1M context opens via --max-model-len (the Hopper sample caps at 262144). SGLang is not merged to main yet (Blackwell needs the MSA kernel at a pinned commit plus nvcc 13.4 minimum - cu130 does not count). The memory ledger: BF16 at roughly 856GB is an inference, not an official figure; 23B active lowers compute, not the VRAM floor - all 428B experts must stay resident, which is exactly why consumer cards are out; total KV cache for 1M context is unpublished. Domestic compute rides FlagOS Day-0 across nine chips (Huawei Ascend included, via SGLang-plugin-FL), numbers flagged as FlagOS official. Thirteen official error notes collected with fixes: HugeTLB 560GiB, B200 OOM, parser spelling, and more. License: MiniMax Community License, commercial attribution plus the 20-million-dollar threshold. One line of reality: 428B weights are out - budget 8 H100s.

Nemotron 3.5 Bolt Local Deployment: 74B on a 24GB Card

NVIDIA released Nemotron 3.5 Bolt 74B-A9B 24GB on October 15, 2026: a compression of the 120B Super VL via Star Elastic budget-conditioned architecture search (expert top-K down from 22 to 9, distilled recovery training over 120B tokens), on a hybrid Mamba-2 and Transformer architecture. The quantization plan is the story - the expert layers that dominate the weights drop to IQ2_XXS at 2.0625 bits per element (about 19GB), Mamba linear layers use NVFP4, and lm_head, embeddings, and attention stay at Q8_0, landing the full model at 22-24GB with accuracy recovered through quantization-aware distillation from the compression stage onward. Four deployment routes, with commands: Ollama (supported since 0.40.5, ollama run nemotron-3.5-bolt, a 25GB download, verify 100 percent GPU via ollama ps); Unsloth (UD-IQ2_XXS about 22GB, UD-IQ1_M about 20.3GB, ctx from a conservative 4096); llama.cpp official recipes (builds after October 15 to recognize nemotron_h_moe; the 28GB recipe runs 65K context with 8-bit KV and 6 draft tokens, the 32GB recipe reaches 262K; llama-server defaults to port 9931 since b11521); and vLLM (Blackwell SM120 with the LiLiCorr drafter for multi-user serving). Two context starting points coexist: 16K-32K for bare weights, 4K for Unsloth whose default drafter eats another 1-2GB. Eight troubleshooting entries cover OOM context cuts, layers spilling to system RAM, the missing --mmproj flag, and the shared 16384 max_tokens budget. OpenMDW-1.1 grants free unrestricted commercial use, leaves outputs unencumbered, requires notices only on redistribution, and carries a patent-retaliation clause. Vendor-reported: GPQA Diamond 86.4, Terminal-Bench 2.1 at 51.0, PinchBench first in class (83.99 TPS on RTX Spark; no official consumer-card figure). The 4GB Bolt E4B followed on October 23.

Xing4.0-29B-Intent Local Deploy: 4B Active, Which Command?

Getting Xing4.0-29B-Intent running hinges on a detail that trips most people up: the official deployment instructions do not live in the intent repository but in the base model repository. The intent README only says to replace the model path with the Xing4.0-29B-Intent weights and set --served-model-name to Xing4.0-29B-Intent, so the full vLLM and SGLang launch commands must be taken from the Xing4.0-29B-A4B base repo. Several flags are hard requirements: --trust-remote-code is mandatory because model_type is the in-house xing4_0 architecture, neither llama nor qwen; --tensor-parallel-size 2 implies at least two cards; --max-model-len 262144 matches the 256K context; --reasoning-parser xing4_0 and --tool-call-parser xing4_0 supply the matching parsers; --speculative-config uses the mtp method; and --gpu-memory-utilization 0.90 with --max-num-seqs 4 is the official example ratio. The weights ship as 41 safetensors shards totaling 62.43 GB, and the only official quantizations are FP8 (33.17 GB, e4m3) and GGUF (a single IQ4_NL file at 17.75 GB) - there is no AWQ, GPTQ, or INT8 safetensors. Minimum VRAM is not published at all; the official line is to configure GPU memory, parallelism, and context length for your environment. What the base repo does spell out: bfloat16 is required because the mHC path errors out on FP16 or quantized weights, --pipeline-parallel-size greater than 1 is not yet supported, the CUDA mHC path needs tilelang installed, and the host needs Docker 24.0 or newer with driver 535 or newer. The Ascend route explicitly wants 910B2 NPUs with at least two cards, with three official quay.io images (none of which bundle the weights). Consumer and CPU paths appear only in the base llama.cpp docs: int4 weights quantized with IQ4_NL come to about 19 GB and fit on a single consumer GPU, with an RTX 3090 24 GB as the example. The SDK is just an httpx HTTP client and is not on PyPI, and structured output comes in two layers - the model's native category plus args array, and SDK-guided JSON Schema constraints through vLLM response_format. The license is Apache-2.0, yet the weight side has no standalone LICENSE file, the repository LICENSE is an unfilled placeholder template, and the original terms of the 61 upstream datasets still need checking.

Five Open Speech Models in One Drop: Ant's MIT-Licensed Ming-3.0

Ant Group's inclusionAI open-sourced the five-model Ming-3.0 speech family under MIT on 2026-10-19/20 (multi-source confirmed; self-host, fine-tune, and commercial use all allowed), covering synthesis, audio understanding and dialogue, and recognition: Ming-3.0-tts-0.5B (0.5B dense, the lightest, deployable on four cards), Ming-3.0-tts-tiny (7.9B total / 1.3B active MoE, long-form multi-speaker podcasts), Ming-3.0-audio-flash (126B total / 6.5B active MoE, audio understanding and dialogue, answering in text or speech), Ming-3.0-audio-tiny (9.3B / 2.7B, efficiency-oriented), and Ming-3.0-asr-tiny (8.6B / 2B, context-aware transcription). Capabilities: TTS with text-described voices, a director mode that controls emotion, stress, and pacing in natural language, voice cloning, seven languages plus two dialects; audio understanding takes long recordings directly with structured output, preserving emotion and speaker identity, plus tool-call generation; ASR adds dialect recognition, hotword injection, and context awareness. Setup (verified against kie.ai and the README): TTS route - git clone github.com/inclusionAI/Ming-3.0-tts, conda env (0.5B: python 3.10, pip install -r requirements.txt, conda install -c conda-forge pynini==2.1.6), run per README; tts-tiny needs its own env with Python 3.10.15, PyTorch/Torchaudio 2.7.1 (cu118), Transformers 4.46.3; audio-flash via downloaded weights or Ling Studio Chat (ling.tbox.cn), 128K context, speech output through the Chat Completions audio.voice field with modalities=["text","audio"]. Hardware: officially tested mainly on datacenter cards - recommended NVIDIA H800 (80GB) or H20 (96GB); 0.5B on four cards is the floor; consumer-GPU feasibility unverified. Day-0 ecosystem: vLLM-Omni and SGLang-Omni adapted. Internal Ant usage (official): asr-tiny in the Afu doctor assistant, Alipay Search, Yixiaobao realtime voice, and security verification. Four limitations stated as-is: no quantized files (no GGUF/AWQ/EXL2), streaming in doubt (SGLang-Omni returns a full waveform even with stream=true, so true first-audio latency is unestablished), mutually exclusive control modes (voice design, emotion, and dialect cannot combine in one call), and unpublished training data. Leaderboard backing (secondary): Artificial Analysis TTS leaderboard (Provider Voices) puts Ming-3.0-tts at 1244, near the top. Division note: voice-cloning-toolkit-comparison-review (2026-10-15) is the review-lane selection piece; this SOP lane covers hands-on setup of the five-model stack - different lanes, different angles.

AI Cross-Session Memory in 10 Minutes: MemHub Hands-On Guide

A hands-on SOP for wiring Tencent Hunyuan's ready-made memory stack into your AI agent, with every command taken from the MemHub README and verified against PyPI. Three install layers: the SDK via pip install hy-memory==2.0.3 (latest at writing time); the CLI via the npm memhub-cli bootstrapper (about 384KB) followed by memhub install to fetch the Runtime; data lives in ~/.one-memory. Platform support at v1.0.3: Linux x64/arm64 Runtime, macOS arm64/x64 App - Windows is not supported (explicit in the README; bolded in the body plus an FAQ, since many of our readers run Windows). Ten CLI commands walked through one by one: serve/start/stop/restart/status/doctor/install/upgrade/uninstall/delete-data, with a first-run flow example (install, then status to verify, doctor for a health check, serve, then connect your agent). Privacy boundary stated as-is: no screen capture, no OCR, no keyboard/mouse event collection, no clipboard reading, no audio recording. Tool count in two versions: the MemHub README lists 5 tools while Tencent's official WeChat article says 10 - the body picks neither and defers to actual CLI output. A standalone preview warning: both Hy Memory 2.0 and MemHub are preview-stage; commands and APIs may change, use with care in production. Pitfalls covered: a fallback route for Windows readers (WSL2 or wait for official support), pinning ==2.0.3 for reproducibility, and backing up ~/.one-memory by copying the directory. Division note: ai-agent-long-term-memory-sop (2026-08-08) covers the build-your-own mem0/Letta route (vector store plus files); this piece is Tencent's ready-made stack for readers who want it running today. Primary search phrase "AI cross-session memory" (S1=2/S2=1/S3=2/S4=3, about 8 points) lives in the title and this summary only, never in the article body.

Claude Code DeepSeek Setup: Nine Env Vars, No Old Tutorials

A hands-on SOP for wiring Claude Code straight into DeepSeek, with every command and variable taken from the official guide (api-docs.deepseek.com/quick_start/agent_integrations/claude_code), reflecting the current configuration after the 1017 checkpoint. Primary search phrase: "Claude Code DeepSeek setup" (S1=3/S2=1/S3=2/S4=3). Migration path with nine environment variables: ANTHROPIC_BASE_URL to https://api.deepseek.com/anthropic, ANTHROPIC_AUTH_TOKEN with your DeepSeek API key, ANTHROPIC_MODEL=deepseek-flash[1m], DEFAULT_OPUS and DEFAULT_SONNET also deepseek-flash[1m], DEFAULT_HAIKU=deepseek-flash, CLAUDE_CODE_SUBAGENT_MODEL=deepseek-flash, CLAUDE_CODE_EFFORT_LEVEL=max, CLAUDE_CODE_AUTO_COMPACT_WINDOW=786432; both PowerShell and Linux/Mac variants are given. Install from scratch in four steps: Node.js 18+ (plus Git for Windows on Windows), npm install -g @anthropic-ai/claude-code, verify with claude --version, set the variables and run claude in your project directory. Model mapping: requests starting with claude-opus route to deepseek-v4-pro (billed at V4 Pro prices, pricier than Flash - a must-read reminder), claude-haiku and claude-sonnet route to deepseek-flash. Web Search works natively but bills on two lines: search requests per the Search API plus extra token costs for summarizing retrieved content. Pricing (official rate page): off-peak 0.15 dollars per million input tokens on cache miss, 0.003 on a hit (one fiftieth of the miss price), 0.60 output; peak (weekdays 01:00-04:00 and 06:00-10:00 UTC) doubles everything; weekends and Chinese public holidays count as off-peak. Five pitfalls: AUTH_TOKEN and API_KEY are not the same thing (mixing them equals no config), old and new variables coexisting can cause auth conflicts (keep only the new ones, per community guides), never commit keys via settings.json (persist them in ~/.bashrc, ~/.zshrc or system environment variables), validate the tool-calling chain with a minimal read-one-file-then-edit-one-file task before real work, and check the live model with /status. Older tutorials still reference retired names like deepseek-chat. Division of labor: deepseek-v4-pro-claude-code-sop is the V4-Pro-0813-era two-variable minimal setup (2026-09), while this piece is the full current official configuration after 1017; for the Harness itself see deepseek-harness-quickstart-sop.

ChatGPT Chrome Setup: Six Steps to Minimal Permissions

A hands-on SOP for letting ChatGPT drive your browser, with every step and UI path taken from the official docs at learn.chatgpt.com. Six steps: first, prerequisites (a paid ChatGPT plan, the desktop app, and one of five supported Chromium browsers - Chrome, Edge, Brave, Opera, Vivaldi; Firefox and Safari are not supported); second, installation (desktop app, Settings > Computer Use, pick your browser, Install through the extension store, approve permissions, then confirm the browser shows Manage; on macOS, Computer Use needs Screen Recording and Accessibility permissions); third, invocation (at-mentions like @Chrome, right-click Ask ChatGPT, Cmd+Shift+. on macOS for the side panel, bringing open tabs or selected text into the chat, and the double-Command Appshot); fourth, choosing among three surfaces (tasks needing your existing signed-in profile go to the extension, localhost and preview pages to the built-in @Browser, dedicated integrations to plugins); fifth, permission hygiene with the four-level prompt (Allow once, Allow for this site, Allow for all sites, Decline) plus domain allowlist and blocklist management shared across all five browsers, noting the docs themselves flag Allow for all sites and browser-history access as elevated-risk; sixth, safety habits (treat page content, selected text and video transcripts as untrusted context, review pages before acting, do not leave the agent resident). Known quirks from the official docs: Opera lacks the side chat but supports desktop workflows, only Chromium browsers are supported, and Chrome tasks run grouped in tab groups. Division of labor with the hotspot piece: that one covers OpenAI's product move; this one covers configuring your own machine and keeping permissions minimal.