> Markdown mirror of https://router.one/blog/new-llm-models-november-2026 for AI assistants and crawlers. Router One is a unified, OpenAI-compatible LLM API gateway.
> Published: 2026-11-11 · Author: Router One Team

# New LLMs, Nov 2026: Claude Fable 5.2, Gemini 4 Carbon

_New on Router One in November 2026: Claude Fable 5.2, Gemini 4 Carbon and GPT Image 2.6 Flare and Sunburst, with model IDs, endpoints and plan coverage._

This is the November 2026 sibling of the [October 2026 new-model guide](https://router.one/blog/new-llm-models-october-2026), written the same way: for the listings and plan changes this post records, observed after that post's wrap-up (the 16:04 UTC catalog read on 2026-10-30) and up to 2026-11-11, what the [live catalog](https://router.one/models) and the plan responses show for the ids involved — which endpoints serve them, whether a plan covers them — and how to send a first request. As of the 2026-11-05 plan response, Grok 4.7 (`grok-4.7`) is in the Mid-tier models tier of the Pro, Max and Ultra plans; the 2026-11-02 plan response listed it in no tier. As of the 2026-11-05 catalog, Router One lists GPT Image 2.6 Flare (`gpt-image-2.6-flare`) and GPT Image 2.6 Sunburst (`gpt-image-2.6-sunburst`). As of the 2026-11-11 catalog, Router One lists Claude Fable 5.2 (`anthropic/claude-fable-5.2`) and Gemini 4 Carbon (`google/gemini-4-carbon`), and the gateway also accepts the two official ids, `claude-fable-5-2` and `gemini-4-carbon`; no plan tier lists either model in the 2026-11-11 plan response.

Listing dates in this post are catalog observation dates, not vendor launch dates, and the [changelog](https://router.one/blog/changelog) records the same catalog and plan changes.

## November at a glance (as of 04:38 UTC on 2026-11-11)

The table lists the listings and plan changes this post records up to the 04:38 UTC catalog read on 2026-11-11, the endpoints that serve each id and whether a plan covers it; live rates are on each model page, not here.

| Model | Catalog id | First seen | Endpoints | Plan coverage (2026-11-11) |
| --- | --- | --- | --- | --- |
| [Claude Fable 5.2](https://router.one/models/claude-fable-5-2) | `anthropic/claude-fable-5.2` | The 2026-11-11 catalog | Messages, Chat Completions | None — wallet, per token |
| [Gemini 4 Carbon](https://router.one/models/gemini-4-carbon) | `google/gemini-4-carbon` | The 2026-11-11 catalog | Chat Completions | None — wallet, per token |
| [GPT Image 2.6 Flare](https://router.one/models/gpt-image-2-6-flare) and [GPT Image 2.6 Sunburst](https://router.one/models/gpt-image-2-6-sunburst) | `gpt-image-2.6-flare`, `gpt-image-2.6-sunburst` | The 2026-11-05 catalog | Images: generations | None — wallet, per image |
| [Grok 4.7](https://router.one/models/grok-4-7) (plan change) | `grok-4.7` | The 2026-11-05 plan response (id listed 2026-09-22) | Responses, Chat Completions | Mid-tier models tier of Pro, Max and Ultra |

As of the 2026-11-11 catalog, Claude Fable 5.1 (`anthropic/claude-fable-5.1`) and Claude Fable 5 (`anthropic/claude-fable-5`) stay listed beside Claude Fable 5.2, and Gemini 4 Argon (`google/gemini-4-argon`) stays listed beside Gemini 4 Carbon. Plan model lists change, and the [pricing page](https://router.one/pricing) shows the live ones with each plan's allowance; a model that no plan tier lists bills from the wallet, per token or per image, whether or not you hold a plan.

## Claude Fable 5.2

Router One lists Claude Fable 5.2 as `anthropic/claude-fable-5.2` as of the 2026-11-11 catalog, and the gateway also accepts Anthropic's own id, `claude-fable-5-2`. The catalog entry lists text and image input, text output and the chat, streaming, tool-calling and vision flags; the context window and the live rates are on the [Claude Fable 5.2 model page](https://router.one/models/claude-fable-5-2). As a Claude-family id it is served on `POST /v1/messages`, the path Claude Code uses, and on `POST /v1/chat/completions`, not on `POST /v1/responses`.

Per Anthropic's Claude Platform release notes (November 9, 2026; checked 2026-11-11), Anthropic launched Claude Fable 5.2 (`claude-fable-5-2`), which it describes as "built for demanding reasoning and long-horizon agentic work", with a 1M token context window and 128k max output tokens on Anthropic's own API. Those are Anthropic's statements; Router One is responsible for the catalog entry and for how the gateway routes and bills it.

No plan tier in the 2026-11-11 plan response lists Claude Fable 5.2, so its calls bill per token from the wallet, plan or no plan. In the same plan response, no tier lists [Claude Fable 5.1](https://router.one/models/claude-fable-5-1), [Claude Fable 5](https://router.one/models/claude-fable-5) or [Claude Opus 5.5](https://router.one/models/claude-opus-5-5) either. The [Claude Fable 5.2 API guide](https://router.one/blog/claude-fable-5-2-api-guide) covers the id, Anthropic's own notes on the model and a first call, and [Claude Fable 5.2 vs Claude Opus 5.5](https://router.one/models/compare/claude-fable-5-2-vs-claude-opus-5-5) renders the spec sheets, live rates and plan coverage side by side.

In Claude Code, Claude Fable 5.2 arrives through the `fable` alias. Per Claude Code's changelog (checked 2026-11-11), from 2.1.257 the alias requests Claude Fable 5.1 (claude-fable-5-1), and from 2.1.321 it requests Claude Fable 5.2 (claude-fable-5-2). As of the 2026-11-11 catalog, Router One lists both, as `anthropic/claude-fable-5.1` and `anthropic/claude-fable-5.2`, and accepts both official ids, so on 2.1.321 or later the alias reaches Claude Fable 5.2 with no override. If you set `ANTHROPIC_DEFAULT_FABLE_MODEL=claude-fable-5-1` on this site's earlier advice (2026-11-10), you can delete that line; keep it if you want the alias to stay on Claude Fable 5.1. Neither Claude Fable 5.2 nor Claude Fable 5.1 is in a plan tier as of the 2026-11-11 plan response, so either way the calls bill to the wallet. The [Claude Code setup guide](https://router.one/blog/claude-code-setup-guide) has the full block and explains each line.

## Gemini 4 Carbon

Router One lists Gemini 4 Carbon as `google/gemini-4-carbon` as of the 2026-11-11 catalog, and the gateway also accepts Google's own id, `gemini-4-carbon`. The catalog entry lists text and image input, text output and the chat, streaming, tool-calling and vision flags; the live rates are on the [Gemini 4 Carbon model page](https://router.one/models/gemini-4-carbon). As a Gemini id it is served on `POST /v1/chat/completions` only; Claude Code sends Anthropic Messages requests to `/v1/messages`, which rejects Gemini ids with HTTP 400, so Claude Code cannot call Gemini 4 Carbon.

Per Google's Gemini API release notes (November 10, 2026; checked 2026-11-11), Gemini 4 Carbon (Google's model id `gemini-4-carbon`) is generally available on the Gemini API. Those are Google's statements; Router One is responsible for the catalog entry and for how the gateway routes and bills it.

No plan tier in the 2026-11-11 plan response lists Gemini 4 Carbon, so its calls bill per token from the wallet, plan or no plan. As of the 2026-11-11 plan response, Gemini 4 Argon is in the Premium models tier of the Pro, Max and Ultra plans. [Gemini 4 Carbon vs Gemini 4 Argon](https://router.one/models/compare/gemini-4-carbon-vs-gemini-4-argon) renders both spec sheets and live rates side by side, and the [Gemini 4 Carbon API guide](https://router.one/blog/gemini-4-carbon-api-guide) covers the catalog id, the endpoint and a first request.

Access from mainland China for the whole Gemini family is on the [Gemini API in China page](https://router.one/gemini-api-china).

## GPT Image 2.6 Flare and Sunburst

As of the 2026-11-05 catalog, Router One lists [GPT Image 2.6 Flare](https://router.one/models/gpt-image-2-6-flare) (`gpt-image-2.6-flare`) and [GPT Image 2.6 Sunburst](https://router.one/models/gpt-image-2-6-sunburst) (`gpt-image-2.6-sunburst`), first seen in the 16:05 UTC read on 2026-11-05: image models with text input and image output, served on `POST /v1/images/generations` and priced per image; the catalog gives no release date. As of the 2026-11-10 catalog, the five GPT Image ids (`gpt-image-2`, `gpt-image-2.5-flare`, `gpt-image-2.5-sunburst`, `gpt-image-2.6-flare` and `gpt-image-2.6-sunburst`) are $0.10 per image. No tier of the 2026-11-11 plan response lists an image model, so each image is charged to the wallet, plan or no plan.

The [GPT Image 2.6 API guide](https://router.one/blog/gpt-image-2-6-api-guide) covers the two ids, OpenAI's own description of Flare and Sunburst and a first call, and the [image generation API page](https://router.one/image-generation-api) covers the image models on the gateway, each with its per-image price on its model page.

## Grok 4.7 in the Mid-tier models tier

As of the 2026-11-05 plan response, [Grok 4.7](https://router.one/models/grok-4-7) (`grok-4.7`) is in the Mid-tier models tier of the Pro, Max and Ultra plans, beside Grok 4.6, Grok 4.5 and the other models that tier lists; the 2026-11-02 plan response listed it in no tier, and the 2026-11-11 plan response still lists it in the Mid-tier models tier. On a plan, its calls count against the Mid-tier models allowance, and calls beyond the allowance bill the wallet at the posted rates; without a plan, every call bills the wallet per token. [Grok 4.7 Build Fast](https://router.one/models/grok-4-7-build-fast) (`grok-4.7-build-fast`) is a separate id and, in the same plan responses, in no plan tier.

As a Grok chat id, Grok 4.7 is served on `POST /v1/chat/completions` and natively on `POST /v1/responses`. The [Grok API in China page](https://router.one/grok-api-china) lists the Grok chat ids and covers access from mainland China, and [Grok 4.7 vs GPT-6 Astra](https://router.one/models/compare/grok-4-7-vs-gpt-6-astra) renders both spec sheets and plan coverage side by side.

## Gemini 3.8 Flash: a guide and a comparison

Gemini 3.8 Flash (`google/gemini-3.8-flash`) is not new to the catalog; what is new after the October post is the documentation. The [Gemini 3.8 Flash API guide](https://router.one/blog/gemini-3-8-flash-api-guide), published 2026-11-05, covers the catalog id, the Chat Completions endpoint, Google's thinking levels and price dates, plan quota and wallet billing and a first request, and [Gemini 3.8 Flash vs Gemini 3.7 Flash](https://router.one/models/compare/gemini-3-8-flash-vs-gemini-3-7-flash) is a new comparison page. As of the 2026-11-11 plan response, Gemini 3.8 Flash is in the Standard models tier of the Pro, Max and Ultra plans, so on a plan its calls count against the Standard models allowance.

## GPT-6 Luna: no catalog change, a fuller guide

GPT-6 Luna (`openai/gpt-6-luna`) and its Azure channel id `azure/gpt-6-luna` are listed as of the 2026-10-30 catalog — the October post records the listing — and both are in the 2026-11-11 catalog. No plan tier lists either in the 2026-11-11 plan response, so their calls bill per token from the wallet, plan or no plan. The [GPT-6 Luna API guide](https://router.one/blog/gpt-6-luna-api-guide) has grown since the October post: Router One serves both ids natively on the Responses API and resolves the bare id `gpt-6-luna` to `openai/gpt-6-luna` (observed 2026-10-31), the guide shows the one-line Codex CLI setting `model = "gpt-6-luna"`, and as of 2026-11-05 it covers how the gateway handles Chat Completions requests to `openai/gpt-6-luna` that carry tools.

## Which endpoint serves which id

Endpoint families are not interchangeable. `POST /v1/chat/completions` serves every chat model in the catalog; `POST /v1/messages` serves the Claude-family and DeepSeek ids; `POST /v1/responses` is served natively for the GPT-family and DeepSeek ids and for the Grok chat ids. A Claude-family id sent to `/v1/responses`, or a chat id outside the Claude and DeepSeek families sent to `/v1/messages`, is rejected with HTTP 400 `invalid_request_error` before any model is called, and the message names the path to use. Image models take the Images API instead. For the ids in this post that means: Claude Fable 5.2 on Messages or Chat Completions, not on Responses; Gemini 4 Carbon and Gemini 3.8 Flash on Chat Completions only; GPT Image 2.6 Flare and Sunburst on `POST /v1/images/generations`, the endpoint their model pages list; Grok 4.7, GPT-6 Luna and `azure/gpt-6-luna` on Responses or Chat Completions. The [OpenAI-compatible API page](https://router.one/openai-compatible-api) covers the request shape, and each model page lists exactly the endpoints it serves.

## How to try one in five minutes

1. **Create a key.** Dashboard → API Keys → Create Key. Keys look like `sk-...`. If this is an experiment, give the key a `maxSpend` cap so a runaway loop's wallet charges stop at a number you chose ([per-key spend caps](https://router.one/llm-cost-tracking)).
2. **Set the base URL.** `https://api.router.one/v1` for OpenAI-compatible clients and SDKs; the chat and image endpoints both sit under it and take the same key.
3. **Send one request per model.**

Gemini 4 Carbon on Chat Completions:

```bash
curl https://api.router.one/v1/chat/completions \
  -H "Authorization: Bearer sk-your-api-key" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "google/gemini-4-carbon",
    "messages": [
      {"role": "system", "content": "Answer in three bullet points."},
      {"role": "user", "content": "What should a retry policy for a payment webhook include?"}
    ]
  }'
```

GPT Image 2.6 Flare on the Images API:

```bash
curl https://api.router.one/v1/images/generations \
  -H "Authorization: Bearer sk-your-api-key" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-image-2.6-flare",
    "prompt": "Illustration of a city street after the rain, soft pastel colors, no text"
  }'
```

Every image the call returns is billed at the per-image unit price on the model page. The [GPT Image 2.6 API guide](https://router.one/blog/gpt-image-2-6-api-guide) shows how to read the response, and the [Images API docs](https://router.one/docs/images/createImageGeneration) list the fields. For Claude Fable 5.2 on `POST /v1/messages`, the [Claude Fable 5.2 API guide](https://router.one/blog/claude-fable-5-2-api-guide) has the request.

4. **Read the trace.** Dashboard → Logs lists both calls with the model, the wallet charge or plan quota each one used, latency and status, so you see what each call drew before you wire either model into your app ([per-request observability](https://router.one/llm-observability)).

Top up with a card or Alipay through one hosted checkout, or with USDT/USDC on six chains (Tron, BSC, Ethereum, Polygon, Base, Arbitrum). No US credit card required.

## FAQ

**What is the model id for Claude Fable 5.2, and does a plan cover it?**
The id is anthropic/claude-fable-5.2, listed as of the 2026-11-11 catalog; the gateway also accepts Anthropic's own id, claude-fable-5-2. Call it on /v1/messages or /v1/chat/completions with a Router One key; as a Claude-family id it is not served on /v1/responses. No plan tier in the 2026-11-11 plan response lists it, so its calls bill per token from the wallet balance at the rates on its model page, plan or no plan. The [Claude Fable 5.2 API guide](https://router.one/blog/claude-fable-5-2-api-guide) covers the first call.

**Which Fable model does Claude Code's fable alias request?**
Per Claude Code's changelog (checked 2026-11-11), from Claude Code 2.1.257 until 2.1.321 the fable alias requests claude-fable-5-1, and from 2.1.321 it requests claude-fable-5-2. As of the 2026-11-11 catalog, Router One lists both (anthropic/claude-fable-5.1 and anthropic/claude-fable-5.2) and accepts both official ids. If you set ANTHROPIC_DEFAULT_FABLE_MODEL=claude-fable-5-1 earlier, delete the line to follow the alias to Claude Fable 5.2, or keep it to stay on Claude Fable 5.1; no plan tier lists either model in the 2026-11-11 plan response.

**What is the model id for Gemini 4 Carbon, and does a plan cover it?**
The id is google/gemini-4-carbon (the gateway also accepts Google's own id, gemini-4-carbon), listed as of the 2026-11-11 catalog and called on /v1/chat/completions with a Router One key; Claude Code's /v1/messages path rejects Gemini ids with HTTP 400. No plan tier in the 2026-11-11 plan response lists it, so its calls bill per token from the wallet, plan or no plan. The [Gemini 4 Carbon API guide](https://router.one/blog/gemini-4-carbon-api-guide) covers the first call.

**Should I use Gemini 4 Carbon or Gemini 4 Argon?**
Both are Chat Completions ids with text and image input in the 2026-11-11 catalog, each with its own rates on its model page. Per Google's What's new in Gemini 4 Carbon page (checked 2026-11-11), Gemini 4 Carbon spends more thinking tokens per turn than Gemini 4 Argon, which raises per-turn token use and latency, while Gemini 4 Argon gives faster responses and lower token spend per turn. No plan tier in the 2026-11-11 plan response lists Gemini 4 Carbon, so its calls bill the wallet. As of the 2026-11-11 plan response, Gemini 4 Argon is in the Premium models tier of the Pro, Max and Ultra plans, so on a plan its calls draw that tier's allowance. Run your own prompts through both and compare each request's wallet charge or plan quota in Dashboard → Logs; [Gemini 4 Carbon vs Gemini 4 Argon](https://router.one/models/compare/gemini-4-carbon-vs-gemini-4-argon) renders both spec sheets and live rates.

**How are GPT Image 2.6 Flare and Sunburst billed?**
Per image, from the wallet. As of the 2026-11-10 catalog, gpt-image-2.6-flare and gpt-image-2.6-sunburst are $0.10 per image each, and no tier of the 2026-11-11 plan response lists an image model, so every image is charged to the wallet, plan or no plan. Both ids, listed as of the 2026-11-05 catalog, take the same POST /v1/images/generations request; the [GPT Image 2.6 API guide](https://router.one/blog/gpt-image-2-6-api-guide) covers how Flare and Sunburst differ in OpenAI's own words.

**Is Grok 4.7 covered by a plan now?**
Yes, on a Pro, Max or Ultra plan. As of the 2026-11-05 plan response, grok-4.7 is in the Mid-tier models tier of all three; the 2026-11-02 plan response listed it in no tier, and the 2026-11-11 plan response lists it in the same tier. Its calls draw the Mid-tier models allowance, and calls beyond the allowance bill the wallet at the posted rates. grok-4.7-build-fast is a separate id and, in the same plan responses, in no plan tier.

**Can I still use Claude Fable 5.1?**
Yes. As of the 2026-11-11 catalog, Router One lists Claude Fable 5.1 (anthropic/claude-fable-5.1) and Claude Fable 5 (anthropic/claude-fable-5) beside Claude Fable 5.2. No plan tier in the 2026-11-11 plan response lists any of the three, so their calls bill per token from the wallet. Call it by its catalog id or by Anthropic's id claude-fable-5-1. In Claude Code 2.1.321 or later, /model claude-fable-5-1 selects it by id (per Claude Code's model configuration docs, checked 2026-11-11), and an ANTHROPIC_DEFAULT_FABLE_MODEL=claude-fable-5-1 line you set earlier keeps the fable alias on it.

## Next steps

- Check the [Claude Fable 5.2](https://router.one/models/claude-fable-5-2), [Gemini 4 Carbon](https://router.one/models/gemini-4-carbon), [GPT Image 2.6 Flare](https://router.one/models/gpt-image-2-6-flare), [GPT Image 2.6 Sunburst](https://router.one/models/gpt-image-2-6-sunburst) and [Grok 4.7](https://router.one/models/grok-4-7) model pages for live rates and capability flags.
- Call Claude Fable 5.2 with the [Claude Fable 5.2 API guide](https://router.one/blog/claude-fable-5-2-api-guide), and compare it with [Claude Opus 5.5](https://router.one/models/compare/claude-fable-5-2-vs-claude-opus-5-5) before you switch.
- Call Gemini 4 Carbon on Chat Completions with the [Gemini 4 Carbon API guide](https://router.one/blog/gemini-4-carbon-api-guide), and compare it with [Gemini 4 Argon](https://router.one/models/compare/gemini-4-carbon-vs-gemini-4-argon) before you switch.
- Generate images with the [GPT Image 2.6 API guide](https://router.one/blog/gpt-image-2-6-api-guide) and the [image generation API page](https://router.one/image-generation-api).
- Read the [Gemini 3.8 Flash API guide](https://router.one/blog/gemini-3-8-flash-api-guide) and [Gemini 3.8 Flash vs Gemini 3.7 Flash](https://router.one/models/compare/gemini-3-8-flash-vs-gemini-3-7-flash), and the [GPT-6 Luna API guide](https://router.one/blog/gpt-6-luna-api-guide).
- See which models each plan tier lists, and each plan's allowance, on the [pricing page](https://router.one/pricing).
- Read the [October 2026 guide](https://router.one/blog/new-llm-models-october-2026) for the previous batch — Gemini 4 Argon, Claude Haiku 5.5, GPT-6.1 Terra and GPT-6 Luna among them.
- Set up Claude Code with the [Claude Code setup guide](https://router.one/blog/claude-code-setup-guide).

## See also

- Canonical page: https://router.one/blog/new-llm-models-november-2026
- LLM API Gateway and Routing: https://router.one/llm-api-gateway
- All blog posts: https://router.one/blog
- Models and per-model token rates: https://router.one/models (markdown: https://router.one/models.md)
- Pricing: https://router.one/pricing
- API docs (markdown): https://router.one/docs.md
- Company facts: https://router.one/facts/company.md