Big-AGI is an open-source workspace for using multiple AI models through chat and other AI functions. It is intended for engineers, founders, researchers, and other users who want to work with AI personas, model comparisons, image generation, voice, documents, and code-related features. The catalogue entries provide commands, instructions, and a skill for working with Big-AGI.
Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/enricoros/big-AGIWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/enricoros/big-agi/update-models-alibaba)<a href="https://agentmods.dev/commands/enricoros/big-agi/update-models-alibaba"><img src="https://agentmods.dev/badge/commands/enricoros/big-agi/update-models-alibaba/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/enricoros/big-agi/update-models-alibaba"><img src="https://agentmods.dev/badge/commands/enricoros/big-agi/update-models-alibaba.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00009 | $0.00895 |
| Opus 5 | $0.00005 | $0.00447 |
| Sonnet 5 | $0.00002 | $0.00179 |
| Haiku 4.5 | $0.00001 | $0.00089 |
Grade A, and why
update-models-alibaba scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
**Fetching the docs:** `curl -L -A "Mozilla/5.0 ..."` returns the full server-rendered page (the tables are in the HTML); strip `<script>` and convert `<tr>/<td>` to text. Both help pages lag new launches by weeks - a mo What it actually says
Update src/modules/llms/server/openai/models/alibaba.models.ts with latest model definitions.
Reference src/modules/llms/server/llm.server.types.ts and src/modules/llms/server/models.mappings.ts for context only. Focus on the model file, do not descend into other code.
Primary Sources:
- Models list: https://www.alibabacloud.com/help/en/model-studio/models
- Price tables (per region, USD for International/Singapore): https://www.alibabacloud.com/help/en/model-studio/model-pricing
- Per-model pages, e.g. https://www.alibabacloud.com/help/en/model-studio/qwen3-8-max (id with
.->-): the only source for max input / max output / max CoT and the per-model cache-hit price, which the price tables omit - Cache billing rules: https://www.alibabacloud.com/help/en/model-studio/context-cache
- Billing Guide: https://www.alibabacloud.com/help/en/model-studio/billing-for-model-studio
Fetching the docs: curl -L -A "Mozilla/5.0 ..." returns the full server-rendered page (the tables are in the HTML); strip <script> and convert <tr>/<td> to text. Both help pages lag new launches by weeks - a model can be live on the API and absent from the models list AND the price tables. Per-model pages also omit whole regions (qwen3.7-max/-flash have no Singapore section while the price table prices them there), so a missing region is not evidence of unavailability. Cache-hit rate is a per-family ratio of input price, stable across regions: read it off any region on the model page and apply it to the Singapore input price (Qwen/Kimi 20%, GLM 25%, deepseek-v4-pro ~8%).
OpenRouter as the tiebreaker (no key needed): curl -s https://openrouter.ai/api/v1/models/qwen/<id>/endpoints - the Alibaba endpoint is Alibaba's own passthrough and carries the exact International list tiers (pricing.overrides[].min_prompt_tokens = the tier boundaries), input_cache_read, max_completion_tokens, max_prompt_tokens, and the modality string. It is often the ONLY published International price for a just-launched model. Note pricing.discount / promo cuts: OpenRouter may show the promotional rate, while this file records the list price (the price tables mark these "List price $X Limited-time N% off").
Fallbacks if blocked:
- Search "alibaba model studio latest pricing", "alibaba latest models", "qwen models pricing", or search GitHub for latest model prices and context windows
Live endpoint (extra signal): The DashScope key in .env.api-keys is ALIBABA_API_KEY. Scan the OpenAI-compatible model list as ground-truth for what's new/available and cross-check the docs above: curl https://dashscope-intl.aliyuncs.com/compatible-mode/v1/models -H "Authorization: Bearer $ALIBABA_API_KEY". Never commit or echo the key. The list carries only id/created/owned_by, but two caps are cheaply probed on /chat/completions: max_tokens: 999999 returns Range of max_tokens should be [1, N] (N differs with enable_thinking on/off), and an oversized prompt returns Range of input length should be [1, N]. Free-tier quota can be exhausted per model, which masks probes. Treat N as the accepted parameter range, not the real output cap - it is permissive (qwen3.7-flash accepts 131072 while the model page and OpenRouter both cap it at 65536); prefer the docs/OpenRouter value when they agree against the probe.
Important:
- Review the full model list for additions, removals, and price changes
- Minimize whitespace/comment changes, focus on content
- Preserve comments to make diffs easy to review
- Flag broken links or unexpected content
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 9d ago First seen · 30 lines · 9 tokens per session scan A 4731b4ca3861
update-models-alibaba is a command published in the GitHub repository enricoros/big-AGI (7,113 stars, last pushed 2d ago), licensed MIT. It adds 9 tokens to every session and 895 once invoked, about $0.0000 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other commands, from other repositories
AGENTS
Per-command agent-optimized usage and structure guide.
forge
You are in FORGE MODE. Execute the following.
apex-architect
Use the @apex-architect agent to help the user with the following architectural or debugging matter: $ARGUMENTS.
echo-analyst
Use the @echo-analyst agent to surface requirement gaps and assumptions for the following: $ARGUMENTS.
agi-improve
Start a self-improvement cycle to enhance capabilities.
agi-research
Trigger autonomous research on a topic to fill knowledge gaps.