Add Nextbit provider with 17 models

[deaquino]

This pull request introduces support for the Nextbit provider and adds configuration files for a wide range of new language models, including detailed metadata and extended capabilities for each model. The changes include the addition of the main provider configuration and TOML files for several new models, specifying their properties, limits, and modalities.

Provider Integration:

• Added providers/nextbit/provider.toml to register the Nextbit provider, including API endpoint, environment variables, and documentation link.

New Model Definitions:

• Added TOML configuration files for multiple models, such as llama3.3:70b, euryale:33-70b, gemma-2:27b-it, ministral:3b, ministral:3-14b, ministral:8b-2512, mythomax:13b, remm-slerp:l2-13b, rocinante:12b, qwen3:14b, qwen3:30b, qwen:3.5-35b, qwen:3.5-9b, deepseek:qwen3-r1-32b, gemma4:26b-a4b, gpt-oss:20b, and microsoft:phi-4. Each file includes model-specific metadata, input/output limits, and supported modalities. [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17]

Model Configuration Enhancements:

• Many models utilize the extends and omit features to inherit from upstream models while omitting cost information, ensuring consistent and simplified configuration management. [1] [2] [3] [4] [5] [6]
• Several models have large context and output limits, supporting advanced use cases and multi-modal inputs (e.g., text, image, video, audio). [1] [2] [3]

Metadata and Capability Specification:

• Each model file provides detailed metadata such as release dates, last updated, family, open weights, and knowledge cutoffs, as well as capability flags (e.g., tool_call, structured_output, reasoning, attachment). [1] [2] [3] [4] [5] [6] [7]

These changes collectively enable seamless integration with the Nextbit provider and make a diverse set of advanced models available for use.

[Copilot]

Pull request overview

Introduces a new nextbit provider entry and adds a set of Nextbit model definition TOML files so these models can be surfaced/queried through the repository’s generated provider/model catalog.

Changes:

• Added providers/nextbit/provider.toml to register the Nextbit provider as an OpenAI-compatible endpoint.
• Added 17 model definition TOMLs under providers/nextbit/models/, mixing fully-authored specs and extends-based inheritance.
• Added a provider logo.svg for Nextbit.

Reviewed changes

Copilot reviewed 18 out of 19 changed files in this pull request and generated 1 comment.

Show a summary per file

File	Description
providers/nextbit/provider.toml	Registers Nextbit provider metadata (env, SDK, API base, docs).
providers/nextbit/logo.svg	Adds a Nextbit provider logo asset.
providers/nextbit/models/rocinante:12b.toml	Adds authored model metadata/limits/modalities for Rocinante 12B.
providers/nextbit/models/remm-slerp:l2-13b.toml	Adds authored model metadata/limits/modalities for ReMM SLERP L2 13B.
providers/nextbit/models/qwen3:30b.toml	Adds authored model metadata/limits/modalities for Qwen3 30B-A3B.
providers/nextbit/models/qwen3:14b.toml	Adds Nextbit model via extends from alibaba/qwen3-14b with limit overrides.
providers/nextbit/models/qwen:3.5-9b.toml	Adds authored multi-modal model metadata/limits/modalities for Qwen3.5 9B.
providers/nextbit/models/qwen:3.5-35b.toml	Adds Nextbit model via extends from alibaba/qwen3.5-35b-a3b with limit overrides.
providers/nextbit/models/mythomax:13b.toml	Adds authored model metadata/limits/modalities for MythoMax 13B.
providers/nextbit/models/mistral:8b-2512.toml	Adds authored model metadata/limits/modalities for an 8B “2512” model.
providers/nextbit/models/ministral:3b.toml	Adds Nextbit model via extends from mistral/ministral-3b-latest with limit overrides.
providers/nextbit/models/ministral:3-14b.toml	Adds authored model metadata/limits/modalities for Ministral 3 14B 2512.
providers/nextbit/models/microsoft:phi-4.toml	Adds Nextbit model via extends from azure/phi-4 with limit overrides.
providers/nextbit/models/llama3.3:70b.toml	Adds Nextbit model via extends from llama/llama-3.3-70b-instruct with limit overrides.
providers/nextbit/models/gpt-oss:20b.toml	Adds Nextbit model via extends from ollama-cloud/gpt-oss:20b.
providers/nextbit/models/gemma4:26b-a4b.toml	Adds Nextbit model via extends from google/gemma-4-26b-a4b-it with limit overrides.
providers/nextbit/models/gemma-2:27b-it.toml	Adds authored model metadata/limits/modalities for Gemma 2 27B IT.
providers/nextbit/models/euryale:33-70b.toml	Adds authored model metadata/limits/modalities for Llama 3.3 Euryale 70B.
providers/nextbit/models/deepseek:qwen3-r1-32b.toml	Adds Nextbit model via extends from alibaba-cn/deepseek-r1-distill-qwen-32b with limit overrides.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[deaquino]

@copilot modify it to include the real logo. https://www.nextbit256.com/Nextbit_horizonta_Blackout@2x.png should be in svg or can be in png? if so, can you generate it? there are any rules for the logo?

square version:
https://www.nextbit256.com/Nextbit_mark_Main@2x.webp

[Copilot]

@copilot modify it to include the real logo. https://www.nextbit256.com/Nextbit_horizonta_Blackout@2x.png should be in svg or can be in png? if so, can you generate it? there are any rules for the log...

I can update it, but I can’t access www.nextbit256.com from this environment right now (DNS resolution fails), so I can’t fetch those files directly.

For this repo, the rule is: provider logos are stored as providers/<provider>/logo.svg and served at /logos/{provider}.svg.

If you can upload the official logo asset(s) to the PR (or share an accessible SVG/raw file), I’ll apply it immediately. If only PNG/WEBP is available, I can convert it into the required logo.svg format in the next commit.

[deaquino]

@copilot modify it to include the real logo. https://www.nextbit256.com/Nextbit_horizonta_Blackout@2x.png should be in svg or can be in png? if so, can you generate it? there are any rules for the log...

I can update it, but I can’t access www.nextbit256.com from this environment right now (DNS resolution fails), so I can’t fetch those files directly.

For this repo, the rule is: provider logos are stored as providers/<provider>/logo.svg and served at /logos/{provider}.svg.

If you can upload the official logo asset(s) to the PR (or share an accessible SVG/raw file), I’ll apply it immediately. If only PNG/WEBP is available, I can convert it into the required logo.svg format in the next commit.

Uploaded

[Copilot]

@copilot modify it to include the real logo. https://www.nextbit256.com/Nextbit_horizonta_Blackout@2x.png should be in svg or can be in png? if so, can you generate it? there are any rules for the...

Done in commit 66095a2 — I replaced providers/nextbit/logo.svg with the uploaded official Nextbit logo asset (using the square mark in the required SVG file format).