Skip to main content
株式会社オブライト
AI2026-07-09

Nous Portal Deep Dive — One OAuth Login for 300+ Frontier AI Models (Claude / GPT / Gemini / DeepSeek / Qwen / Kimi / GLM / MiniMax / Grok) Plus a Tool Gateway (Web Search, Image Generation, TTS, Browser Automation), the Official Backend for Hermes Agent Launched 2026-04-27, Free $0 / Plus $20 With $22 in Rollover-Enabled Credits

Nous Research's Nous Portal launched on 2026-04-27 as an AI-gateway subscription. A single OAuth login unlocks 300+ frontier, agentic-oriented models (Claude, GPT, Gemini, DeepSeek, Qwen, Kimi, GLM, MiniMax, Grok, and more), plus a Tool Gateway (web search, image generation, TTS, browser automation) and Nous Chat. Designed as the official backend for Nous Research's Hermes Agent, it removes per-vendor API-key juggling. Pricing: Free $0/month ($0.10 credit) and Plus $20/month (300+ models, hosted tools, $22 worth of monthly credits with rollover). API access with API Docs enables programmatic integration through an OpenAI-compatible API on a monthly-credit model. Strategic significance: the AI-side equivalent of the integration trend from our "just deploy everything to Cloudflare" column — a "run everything through Nous Portal" approach that avoids per-vendor lock-in and gives fast access to the newest models, with rollover credits making cost management manageable. Caveats: dependency on Nous Portal itself, coupling with the Hermes Agent ecosystem, and data-protection terms remain those of each underlying model vendor at the enterprise contract layer.


TL;DR — What Nous Portal Is

Nous Portal, released by Nous Research on 2026-04-27, is a subscription AI gateway — "300+ models + Tool Gateway + Nous Chat" behind a single OAuth login.

Four takeaways:

1. 300+ frontier models — Claude / GPT / Gemini / DeepSeek / Qwen / Kimi / GLM / MiniMax / Grok, etc.
2. Tool Gateway built in — web search, image generation, TTS, browser automation
3. Official backend for Hermes Agent — Nous's own agent framework
4. Clear pricing — Free $0 ($0.10 credit) / Plus $20/month ($22 credit, rollover)

The Multi-Model AI Gateway, Realized

The 2025–2026 pain: developers want to mix Claude / GPT / Gemini / DeepSeek / Qwen / Kimi, but each vendor is a separate API key, billing model, and SDK. Management cost adds up.

Nous Portal's answer:

- One OAuth login for 300+ models
- OpenAI-compatible API normalizes SDK differences
- Unified credits billing across every model, with rollover for leftovers
- Tool Gateway for web search / image generation / TTS / browser automation on the same key

Difference from LiteLLM / OpenRouter: Nous Portal is tightly integrated with the Hermes Agent framework. It optimizes for agent workflows rather than being a purely generic gateway.

The 300+ Model Roster

Major vendors:

VendorRepresentative modelsRelated columns
AnthropicClaude Sonnet 5 / Opus 4.8 / Haiku 4.5 / Fable 5Sonnet 5, Fable 5 promo
OpenAIGPT family / o-series
GoogleGemini / Gemma 4Gemma 4
DeepSeekV4 Pro etc.Local LLM 2026-06
Alibaba Qwen3.6-35B / 3.5 seriesQwen 3.6 Uncensored
Moonshot KimiK2.7-CodeKimi K2.7
Z.aiGLM-5.2Local LLM 2026-06
MiniMaxM3
xAIGrok series
Nous internalHermes 4 series(this column)

How you reach 300+: ~10 major vendors × multiple generations × derivatives (fine-tunes, quantizations).

The Tool Gateway — More Than Just LLMs

Beyond LLMs, adjacent AI services are unified:

- Web search — real-time info retrieval
- Image generation — DALL-E / Stable Diffusion / Midjourney-family
- TTS — speech synthesis
- Browser automation — web control, scraping

All the parts an agent needs, delivered through one Nous Portal key. Consolidates territory that used to require SerpAPI + Replicate + ElevenLabs + Playwright.

Pricing

Free — $0/month:
- $0.10 monthly credit
- Access to all 300+ models
- Evaluation and learning

Plus — $20/month:
- $22 worth of monthly credit ($20 in, $22 out — effectively 10% off)
- Rollover — unused credit carries forward
- 300+ models plus hosted tool usage
- Recommended tier for Hermes Agent

Billing structure: each model deducts its standard token pricing from your Portal credit balance. Claude Opus 4.8 costs 100× the credit of a Qwen 3.5-3B. Cost of model choice becomes visible in the Portal UI.

Tight Integration With Hermes Agent

Nous Portal is the official backend for Hermes Agent:

- The Hermes Agent desktop app / CLI logs in with Nous Portal
- Model choice, tool calls, and history are centralized in the Portal
- Priority access to Nous's own Hermes 4 series

Positioning: against vendor-specific agents (Claude Cowork, Devin, Cursor), Hermes Agent + Nous Portal offers a vendor-neutral, multi-model agent substrate — the favorite path for the open-weights-friendly crowd.

API Access — OpenAI-Compatible

API Docs provide an OpenAI-compatible API:

python
from openai import OpenAI

client = OpenAI(
    api_key='<YOUR_NOUS_PORTAL_KEY>',
    base_url='https://portal.nousresearch.com/v1',
)

response = client.chat.completions.create(
    model='claude-sonnet-5',  # or 'gpt-5', 'gemini-2.5', 'qwen3.6-35b', ...
    messages=[{'role': 'user', 'content': 'Hello!'}],
)

Existing OpenAI SDKs work as-is — swap the base URL and the key and you have 300+ models. Compatible with LangChain / LlamaIndex / Haystack ecosystems.

Strategic Significance — the AI-Side "Run Everything Through Nous Portal"

This is the AI-layer complement to our "just deploy everything to Cloudflare" column:

- Infra: eight services unified via "everything on Cloudflare"
- AI: 300+ models + tools unified via "everything through Nous Portal"

Benefits:
- No per-vendor API-key juggling
- Frictionless model switching
- Unified credits for cost visibility
- Rollover reduces waste
- Direct pipe into the Hermes Agent ecosystem

Most solo developers and small startups fit inside Plus at $20/month. Enterprise tiers and contracts are expected to expand.

Caveats and Warnings

(1) Dependency on Nous Portal itself: convenient consolidation comes at the price that an outage takes everything down. Consider maintaining direct vendor accounts as a DR / fallback.

(2) Data-protection terms remain per-vendor: prompts still travel to Anthropic / OpenAI / Google under each vendor's terms. For enterprise, direct vendor contracts may offer stronger protection.

(3) Time-to-new-model: expect a few days to weeks between a vendor's release and Portal exposure. If you need day-one access, contract directly.

(4) Credit spend forecasting: 100× price gaps across the 300+ models mean monthly budgets can blow up. Model-selection discipline and usage alerts required.

(5) Coupling with the Hermes Agent ecosystem: sometimes Hermes optimization wins over generic neutrality. If you want pure generic gateway behavior, OpenRouter / LiteLLM are also on the table.

Recommended Actions

Solo developers: Plus $20/month is likely enough. Great fit for use cases that hop between Claude / GPT / Gemini / Qwen / Kimi frequently.

Agent developers: use Hermes Agent + Nous Portal to prototype multi-model, multi-tool agents quickly, with the OpenAI-compatible API to hook into your existing stack.

Enterprise adopters: verify data-protection contract terms, and run direct vendor contracts in parallel for sensitive workloads.

Bottom Line

Nous Portal is the late-2026 landing point for the multi-vendor AI gateway — the AI-layer symmetry to the "just Cloudflare it" trend. 300+ frontier models + Tool Gateway + Nous Chat under one OAuth login, and Plus at $20/month covers most solo-developer and small-startup AI needs. As the official backend for Hermes Agent, it also stakes out the "neutral multi-vendor agent substrate" position against vendor-specific agents like Claude Cowork, Devin, and Cursor.

Related services from us — AI consulting, software development, and OpenClaw setup. For enterprise multi-model AI-substrate design or Hermes Agent + Nous Portal environment setup, get in touch.

References

Feel free to contact us

Contact Us