Global AI desk

AI rankings

Model and product leaderboards. Cite named sources; rankings change fast.

Currently this platform charges no fees

This platform publishes Made in China factory and trade content. Currently this platform charges no fees and holds no escrow. No weapons, military goods, counterfeit, sanctions evasion, vice, or content involving minors. Not a licensed auction house, travel agency, bidding agent, broker, bank, insurer, or customs broker. Nobody promises clearance. Submit a complaint · Community rules

Named desks only. Keep phones and private accounts off posts. Pay against named files. Ads must be labelled.

Email is optional. Anonymous posts cannot be edited or deleted. Register in Member center to change or remove your posts.

Post (Required)

Body (Required)

Optional

Mall, secondhand and other desks: common fields (category, SKU, price, stock, condition…) are optional — tap chips to add, or add all at once. Not a fixed required form. Weapons, medicines, counterfeit, military dual-use, sanctions evasion, gambling, and content involving minors are refused. You can publish with title and body only.

Submit a complaint

Add optional fields (mall/secondhand include the sample listing fields)

Images (Optional)

Videos (Optional)

Up to 12 images (8MB each) and 8 videos (40MB each). Needs review before it is shown. Currently this platform charges no fees.

Email (optional)

Email is optional. Anonymous posts cannot be edited or deleted. Register in Member center to change or remove your posts.

Open-weights AI ranking notes — DeepSeek, GLM, Qwen, MiMo, Gemma, Llama-class (2026)
Currently this platform charges no fees
Category [optional]: AI rankings / open weights Cluster [optional]: Global Open-weights desks matter for factories that need on-prem, air-gap, or custom fine-tunes. Late-2026 LMArena names often include: DeepSeek-V4 / V4.1-Flash (MIT), GLM-5.x (Z.ai, MIT), Qwen3.x open releases (Apache 2.0 on many sizes), MiMo-V2.x (Xiaomi, MIT on Pro/Flash variants), Gemma 4 (Google, Apache 2.0), Meta Llama / Muse open cousins where published, Hy3 (Tencent, Apache 2.0), MiniMax community licenses. Buyer rules: read the exact license file, not the marketing word “open”. Check commercial use, patent grants, and whether “Max/Pro” API siblings differ from downloadable weights. Benchmark your own domain prompts on your GPU/NPU before locking an OEM deal. Pair with AI models board for supply RFQs. Label ads.
Anonymous · 2026-10-31 22:27
Submit a complaint
Global AI text ranking — LMArena Text Arena snapshot (Oct 2026)
Currently this platform charges no fees
Category [optional]: AI rankings Cluster [optional]: Global Source: LMArena Text Arena (arena.ai/leaderboard/chat/text), snapshot ~27 Oct 2026 — ~9.0M votes, ~419 models. Scores move weekly; always re-check the live board before RFQs. Top 30 (Elo-style, proprietary unless noted): 1. Gemini 4 Carbon (High) — Google — ~1528 (preliminary) 2. Gemini 4 Argon (High) — Google — ~1520 — 1M context — ~$2/$10 per M tokens 3. Claude Opus 5.5 (High) — Anthropic — ~1509 — 1M — ~$4/$20 4. Muse Wonder (Max) — Meta — ~1507 — 1M — ~$4/$20 5. Claude Fable 5.1 (Max) — Anthropic — ~1504 — 1M — ~$10/$50 6. Claude Fable 5 (High) — Anthropic — ~1504 7. Claude Opus 4.6 (High) — Anthropic — ~1503 8. Claude Opus 4.7 (High) — Anthropic — ~1501 9. Claude Opus 4.6 — Anthropic — ~1496 10. Gemini 3.8 Flash (High) — Google — ~1496 — ~$0.75/$3.75 11. Muse Spark 1.3 (Max) — Meta — ~1495 12. Claude Opus 4.7 — Anthropic — ~1494 13. Muse Spark 1.2 (xHigh) — Meta — ~1492 14. Claude Opus 5 (High) — Anthropic — ~1492 15. Claude Opus 5 (Max) — Anthropic — ~1491 16. Muse Spark 1.1 — Meta — ~1490 17. Gemini 3.7 Flash (High) — Google — ~1490 18. Muse Spark — Meta — ~1489 19. Kimi K3 (Max) — Moonshot — ~1488 20. Gemini 3.1 Pro Preview — Google — ~1487 21. Gemini 3 Pro — Google — ~1485 22. GPT-5.6 Sol (xHigh) — OpenAI — ~1485 — ~$4/$20 23. Gemini 3.6 Flash (High) — Google — ~1483 24. Qwen3.8-Max — Alibaba — ~1482 — ~$2/$6 25. Claude Opus 4.8 (High) — Anthropic — ~1482 26. GPT-5.5 (High) — OpenAI — ~1482 — 1.1M — ~$5/$30 27. GPT-6.1 Sol (Max) — OpenAI — ~1481 28. GLM-5.3 (Max) — Z.ai (MIT) — ~1481 — ~$1.40/$4.40 29. MiMo-V2.6-Pro — Xiaomi (MIT) — ~1480 — ~$0.43/$0.87 30. Gemini 3.5 Flash (High) — Google — ~1478 Notable 31–50: Claude Sonnet 5.5, GPT-5.5, GLM-5.3-Flash, DeepSeek-V4.1-Flash (MIT), GPT-5.2 Chat, GPT-6 Astra, Grok 4.20 Beta 1 (xAI), Qwen3.7-Max-Preview, ERNIE 5.1 (Baidu). Buyer checklist: exact model id + thinking mode, region, data residency, price per 1M tokens, open vs closed weights. Do not treat Elo as a purchase contract. Label ads. No prepaid to strangers. Live board: https://arena.ai/leaderboard/chat/text
Anonymous · 2026-10-31 22:27
Submit a complaint
AI rankings by use case — chat, coding, image, video, voice, agents (2026 desk map)
Currently this platform charges no fees
Category [optional]: AI rankings Cluster [optional]: Global Not one Elo number. Match the job: Chat / reasoning (closed API, late 2026 desk view): Gemini 4 family, Claude Opus / Fable, Muse Wonder/Spark, GPT-5.x / GPT-6 Sol, Kimi K3, Qwen3.8-Max, GLM-5.3, DeepSeek-V4.x, Grok 4.x. Cite LMArena or named evals with date. Coding / IDE: Cursor (AI-native editor), Claude Code / Anthropic coding modes, GitHub Copilot, Windsurf / similar agents. Require VPC/on-prem options, secret redaction, audit logs, IP terms on generated code. Image gen: Midjourney (aesthetic), Flux / Black Forest Labs (photoreal + open Dev), GPT Image / DALL·E line, Ideogram (text-in-image), Adobe Firefly (enterprise brand kits), Stable Diffusion / SDXL / community forks (self-host). Video: Google Veo (photoreal + native audio), OpenAI Sora, Runway Gen (edit control), Kling (value / I2V), Pika, Luma Dream Machine. State length, watermark, commercial license. Voice: ElevenLabs, OpenAI TTS/Realtime, Azure Speech, Google Cloud TTS, Cartesia, Fish Audio — require written voice likeness consent. Agents: OpenAI Agents / Assistants stack, Anthropic computer-use / tool-use, Google Agent frameworks, open LangGraph/Crew-style stacks — human approval on payments & contracts. Always: named vendor + model id + region + price. Label advertising. No deepfake fraud. No credential theft. Detail cards live under AI models and each use-case board on GlobeVill.
Anonymous · 2026-10-31 22:27
Submit a complaint
AI rankings — how to read leaderboards without hype
Currently this platform charges no fees
Category [optional]: AI rankings Cluster [optional]: Global Leaderboards (LMSYS / Chatbot Arena style, MMLU-class evals, vendor scorecards, app-store charts) are snapshots. Always record: date, benchmark name, open vs closed weights, context length, and who paid for the run. For buyers: ask for the exact model id, region of inference, data residency, and price per 1M tokens — not only a rank number. For sellers: publish named eval links; do not claim “#1 forever”. No prepaid to strangers. Label advertising.
Anonymous · 2026-10-31 08:11
Submit a complaint

Named dest · FCL, single unit, or daigouSubmit a demandDest, FCL / single unit / daigou, SKU. Desk is emailed. Optional messenger, email, and phone stay off the public thread.

Advertising

AI rankings ads

Labelled ads on this column. Must say advertising.

View all