LLM Reference

BAGEL 7B

Released
2025-05-20
Last refreshed
2026-06-21
Status
Researched 1d ago
Open sourceCommercial use: permittedMultimodalVision

BAGEL 7B is a released vision model with open-source; evaluate it while provider pricing coverage matures.

Use it for

  • Teams evaluating vision
  • Workloads that can use a 33k context window

Do not use it for

  • Cost-sensitive launches that need sourced token pricing
  • Strict JSON or tool-calling flows
  • Teams that need a tracked hosted API route today
Specifications
Family
BAGEL
Released
2025-05-20
Context
33k
Parameters
7B
Architecture
Mixture of Transformers
Openness
Open source
License
Apache 2.0OSI-approvedCommercial use: permitted
Created by

TikTok data enhances AI realism

Beijing, China
Founded 2012
Website
Pricing

No tracked provider token pricing is available yet.

About

BAGEL-7B-MoT (Big Advanced Generalized Embodied Learner) is ByteDance Seed's open-source unified multimodal model with 7B active parameters (14B total MoE). Built on Qwen2.5-7B-Instruct with dual encoders for pixel-level and semantic-level features, plus a 32K text backbone context from the Hugging Face config. Supports text conversation, visual reasoning, text-to-image generation, image editing, multiview synthesis, and world navigation. Trained on trillions of interleaved multimodal tokens. Apache 2.0 license, open weights on Hugging Face. No hosted API available as of research date.

BAGEL 7B is an open-source model in the BAGEL family. The structured metadata tracks a 33k-token context window and multimodal input. No headline benchmark score is tracked for BAGEL 7B yet.

Top use-case fit

Vision

Included by capability and metadata signals in the decision map.

Provider price ladder

No tracked provider token pricing is available for this model yet.

Capabilities

VisionMultimodal

Benchmark peer barsfor Vision

No task-mapped benchmark peers are available for this model yet.

Migration checks

No linked migration route is available for this model yet.