| name | minimax-cli |
| description | MiniMax multimodal via mmx-cli — TTS, video, music, image, vision, search. Image generation is PRIMARY choice for Malay/SEA phenotype and realism-critical surfaces. |
| version | 2.0.0 |
| tags | ["minimax","tts","video","music","image","vision","multimodal","malay-phenotype","image-generation"] |
| metadata | {"hermes":{"category":"creative","requires":["mmx-cli"],"related_skills":["token-plan-image","lightweight-image-generation"]},"forge_policy":"/root/A-FORGE/forge_work/2026-07-20/model-selection-policy.md"} |
MiniMax CLI (mmx-cli)
MiniMax multimodal capabilities via mmx CLI. Token Plan subscription required.
Prerequisites
npm install -g mmx-climmx auth login --api-key <sk-cp-key>- Region auto-detected from key (global/cn)
Commands
| Capability | Command | Example |
|---|
| Text chat | mmx text chat --message "..." | mmx text chat --message "Explain NPV" |
| TTS | mmx speech synthesize --text "..." --out voice.mp3 | mmx speech synthesize --text "Hello" --out hi.mp3 |
| Image gen | mmx image generate --prompt "..." | mmx image generate --prompt "sunset ocean" --aspect-ratio 16:9 |
| Video gen | mmx video generate --prompt "..." | mmx video generate --prompt "cat at sunset" |
| Music gen | mmx music generate --prompt "..." | mmx music generate --prompt "jazz summer" --out jazz.mp3 |
| Music gen (lyrics) | mmx music generate --prompt-file tags.txt --lyrics-file lyrics.txt --out song.mp3 | See Music Generation Workflow below |
| Vision | mmx vision describe --file image.png | mmx vision describe --file photo.jpg |
| Web search | mmx search query --query "..." | mmx search query --query "oil price today" |
| Quota | mmx quota | Check remaining Token Plan balance |
| Auth status | mmx auth status | Verify login + region |
Global flags
--api-key <key> — override auth--region global|cn — force region--output json|text — output format--timeout <seconds> — request timeout (default 300)--non-interactive — CI/agent mode (no prompts)
Output
Files saved to minimax-output/ in cwd. When using from Hermes, display media directly in output.
Quota
- Monthly Max: ~5.1B M3 tokens, 3 video/day, 21 video/week
- General models + video + speech + music + image share one quota bar
- 5-hour rolling window + weekly window
- TTS free for limited time (doesn't consume quota)
🔥 Image Generation — PRIMARY for Malay/SEA + Realism
Policy: /root/A-FORGE/forge_work/2026-07-20/model-selection-policy.md
Verdict: MiniMax image-01 is the DEFAULT for federation image generation. Pollinations only for free prototyping.
Basic Usage
mmx image generate --prompt "your prompt" --aspect-ratio 1:1 --non-interactive
Aspect ratios: 1:1 (default), 16:9, 9:16, 4:3, 3:4
Critical pitfall: --output flag is IGNORED by mmx image generate. Files always save as image_001.jpg, image_002.jpg, etc. in the current working directory (not minimax-output/). After generation, find with ls -lt | head and cp to your desired path. Proven 2026-07-20.
🧬 Phenotype
MiniMax image-01 has the strongest SEA/Malay phenotype reading of all available models. When generating images of people:
| Model | SEA Phenotype | Realism | Use Case |
|---|
| MiniMax image-01 | ⭐⭐⭐ Strong | ⭐⭐⭐ Studio-grade | Malay/SEA prompts, realism-critical |
| Qwen image-2.0 | ⭐⭐ Moderate | ⭐⭐ Good | Generic, non-phenotype |
| Pollinations | ⭐ Weak | ⭐⭐ Decent | Free drafts only |
Prompt decomposition for Malay slang:
| Slang | Explicit decomposition |
|---|
abang sado | male, Southeast Asian Malay, muscular build, shirtless, fitness, gym or studio |
Melayu | Southeast Asian Malay ethnicity, natural skin texture, dark hair, brown eyes |
amoi | female, Southeast Asian Chinese/Malay phenotype, young adult |
mat rempit | male, Malay, young, motorcycle, street, urban Malaysia |
Rule: Never rely on the model inferring ethnicity from slang alone. Always add explicit "Southeast Asian Malay" or equivalent phenotype tokens.
🛡️ Safety
When prompt contains shirtless, abang sado, bodybuilding, gym, fitness:
- Default context: gym, studio, outdoor fitness — NOT bedroom, private, intimate
- Pose: physique display, athletic, flexing — NOT sexualized, suggestive
- Framing: full body or torso, fitness lighting — NOT cropped, intimate angles
- Add explicit context: "in a gym", "studio lighting", "fitness photography"
Both MiniMax and Pollinations enforce NSFW filters. MiniMax provides cleaner, more professional fitness-aesthetic results.
⚖️ Contrast
Same prompt shirtless abang sado, Malay, realistic, studio lighting:
| Dimension | MiniMax image-01 | Pollinations |
|---|
| Resolution | 1024×1024 ✅ | 768×768 |
| File size | 184KB ✅ | 73KB |
| Malay phenotype | Strong SEA reading ✅ | Ambiguous/Westernized |
| Realism | Studio-grade, natural ✅ | AI-exaggerated, plastic |
| Prompt understanding | "Abang sado" nailed ✅ | Generic buff guy |
| Cost | Token Plan quota | Free |
Verdict: MiniMax wins clean. For any prompt where Malay/SEA phenotype or realism matters, MiniMax is mandatory.
🎤 Speech
When Arif requests voice messages (TTS), use this fallback order:
- Built-in
text_to_speech tool — uses OpenAI by default, fast, good quality. Fails on quota (429).
edge-tts CLI — free, no API key, good Malay voice. Install: pip install edge-tts --break-system-packages
edge-tts --text "your text" --voice ms-MY-OsmanNeural --write-media /tmp/tts_output.mp3
edge-tts --text "your text" --voice ms-MY-YasminNeural --write-media /tmp/tts_output.mp3
Send with MEDIA:/tmp/tts_output.mp3 in response.mmx speech synthesize — MiniMax TTS, quota-based but high quality.
When user says "voice" or "TTS": try built-in first → if 429 → fall back to edge-tts immediately (no need to ask).
🎵 Music
Full song generation from lyrics + genre tags. Two input files required:
1. Tags file (tags.txt) — comma-separated genre, mood, instrument tags, no spaces after commas:
traditional malay folk,acoustic guitar,gamelan,kompang,joyful,warm,nostalgic,female vocal
2. Lyrics file (lyrics.txt) — bracketed structural tags:
[Intro]
[Verse]
Your lyrics here...
[Chorus]
Chorus lyrics...
[Bridge]
Bridge lyrics...
[Outro]
3. Generate:
mmx music generate \
--prompt-file tags.txt \
--lyrics-file lyrics.txt \
--out song.mp3 \
--non-interactive
Output: MP3, auto-selects model (music-2.6 as of 2026-07). ~4.2MB for a ~3min song, 256kbps stereo 44.1kHz.
Quick prompt-only (no lyrics):
mmx music generate --prompt "jazz piano,relaxing,lo-fi" --out lofi.mp3
Workflow for Telegram delivery:
- Write tags to
/tmp/song_tags.txt
- Write lyrics to
/tmp/song_lyrics.txt
- Run
mmx music generate with --out /tmp/song.mp3
- Verify with
ls -lh /tmp/song.mp3 && file /tmp/song.mp3
- Send with
MEDIA:/tmp/song.mp3 in response
Auth key: source /root/.secrets/vault.env → MINIMAX_API_KEY (sk-cp- prefix, Token Plan).
Cultural/Traditional Song Research Protocol
When user requests a song in a specific cultural/traditional style (e.g., "Kaparinyo", "dondang sayang", "keroncong"):
- Research BEFORE generating. Search for the song's actual origins, regional variant, and musical characteristics. Many folk songs have regional variants with different lyrics and instrumentation.
- Use authentic lyrics — not generic placeholder lyrics. Search for the actual traditional text (often in regional language, not standard Malay/Indonesian).
- Match instrumentation to tradition — e.g., Kaparinyo is Gamad style (violin, accordion, Portuguese guitar, gandang drums), not gamelan.
- Distinguish similar songs — "Kaparinyo" (Minangkabau/Gamad from West Sumatra) ≠ "Burung Kakak Tua" (Ambon/Moluccas). Research prevents conflation.
Audio Evaluation
Two tools for evaluating generated music:
scripts/akal_somatic_scoring.py — ffmpeg-only audio scoring. Three checks: PULSE (temporal fidelity), SOUL (cultural alignment), FLOW (entropy equilibrium). No librosa needed. Returns JSON verdict (SEAL/SABAR/HOLD).templates/cultural_manifold.json — Reusable schema for culturally-grounded music generation. Defines identity, structural priors, harmonic constraints, timbre palette, lyric constraints, and evaluation thresholds. Copy and modify for other traditions (Zapin, Asli, Dondang Sayang, etc.).
Music Evaluation Pipeline
Full evaluation toolkit at /root/music-eval/ — genre scoring, somatic analysis, paradox engine, motif memory, Telegram signal observer. See references/music-eval-pipeline.md for architecture, CLI usage, and A-FORGE integration path. A-FORGE already has paradox-engine/models.py with 16-dim somatic vectors — see references/aforge-paradox-engine.md.
Related References
references/jina-reader-medium.md — Reading Medium/Cloudflare-protected articles via Jina Readerreferences/cross-organ-wiring-pattern.md — Wiring external engines into arifOS kernel (enforcement gates)
👁️ Vision
source /root/.secrets/vault.env
mmx auth login --api-key "$MINIMAX_API_KEY" --non-interactive
mmx vision describe --file /path/to/chart.jpg --non-interactive 2>&1
What it can read from a trading chart:
- Price levels (current, high, low, order levels)
- Chart pattern (downtrend, consolidation, flag, pennant)
- Timeframe (H1, H4 visible from the chart header)
- Support/resistance zones
- Pending orders (buy/sell limits, stop losses)
⚠️ Edge Cases
- Quota exhausted (429) — vision fallback chain. (1) Anthropic API if
ANTHROPIC_API_KEY has credits; (2) MiMo API if MIMO_API_KEY has Token Plan credits; (3) tesseract OCR as last resort.
- 401 after login → set region manually:
mmx config set --key region --value global
mmx image generate --output is ignored — files save as image_001.jpg in current working directory.- Key prefix
sk-cp- = Token Plan (subscription), not pay-as-you-go
- Video is async — poll with
mmx video task get --task-id <id>, then download
- Old SSE MCP servers (minimax-media :18090, minimax-code :18091) are DEAD. Use mmx-cli or stdio minimax-coding-plan-mcp instead
- Output files go to
minimax-output/ in cwd (except image — goes to cwd root)
- "Gemini proposes, Hermes builds" — BUILD what Gemini proposes, don't just critique
Forged: 2026-07-09 · Upgraded: 2026-07-20 (Image Generation + Malay Phenotype section)
Policy: /root/A-FORGE/forge_work/2026-07-20/model-selection-policy.md
DITEMPA BUKAN DIBERI
