Hy4 flash is our most cost-effective Hy model to date, featuring 220B total parameters (including 64B Engram parameters) with just 12B active parameters per token. It natively accepts text and image inputs, generates text, and supports a context window of up to 1M tokens.
Alternatives to Hy4-flash-base
120 open-weight models that do the same job as Hy4-flash-base (219B vision), ranked by how close they are in task and size, then by popularity and how recent they are. Licence and VRAM are shown for each.
Kimi K2.5 is an open-source, native multimodal agentic model built through continual pretraining on approximately 15 trillion mixed visual and text tokens atop Kimi-K2-Base.
1.03TA33.3B256K ctx modified-mit389.7K 2.9KJan 2026- Same category
- Larger (1.03T vs 219B)
- Commercial use OK
The advanced capabilities of the ERNIE 4.5 models, particularly the MoE-based A47B and A3B series, are underpinned by several key technical innovations:
424BA47B128K ctx Apache 2.0428 91Jun 2025- Same category
- Similar size (424B vs 219B)
- Commercial use OK
The package downloads and caches the engine and weights on first use. This repository provides mirage.cact, mirage.safetensors and the model config.
?BApache 2.056.3K 550Nov 2026- Same category
- Commercial use OK
Following the widespread community adoption of the Qwen3.5 and Qwen3.6 series, we are pleased to introduce Qwen3.8, the most capable generation in the Qwen open-model family to date.
27.8B256K ctx Apache 2.06.3M 20.8KAug 2026- Same category
- Smaller (27.8B vs 219B)
- Commercial use OK
- Widely used
Gemma is a family of open models built by Google DeepMind. Gemma 4 models are multimodal, handling text and image input (with audio supported on E2B, E4B, and 12B) and generating text output. This release includes open-weights models in both pre-trained and instruction-tuned variants.
31.3B256K ctx Apache 2.06M 4.4KMar 2026- Same category
- Smaller (31.3B vs 219B)
- Commercial use OK
- Widely used
Mistral Large 4 is our largest and most capable model to date: a 1 trillion-parameter natively multimodal Mixture-of-Experts model with 49 billion active parameters per token (52 billion including embeddings and output layers), trained from scratch on 3,800 NVIDIA Grace Blackwell GPUs in Mistral's…
1TA52B1M ctx Custom licence20.6K 1.7KSept 2026- Same category
- Larger (1T vs 219B)
The advanced capabilities of the ERNIE 4.5 models, particularly the MoE-based A47B and A3B series, are underpinned by several key technical innovations:
424BA47B128K ctx Apache 2.0245 26Jun 2025- Same category
- Similar size (424B vs 219B)
- Commercial use OK
The advanced capabilities of the ERNIE 4.5 models, particularly the MoE-based A47B and A3B series, are underpinned by several key technical innovations:
424BA47B128K ctx Apache 2.0228 68Jun 2025- Same category
- Similar size (424B vs 219B)
- Commercial use OK
In the past five months since Qwen2-VL’s release, numerous developers have built new models on the Qwen2-VL vision-language models, providing us with valuable feedback. During this period, we focused on building more useful vision-language models.
73.4B125K ctx Qwen License38.3K 668Jan 2025- Same category
- Smaller (73.4B vs 219B)
Over recent months, we have intensified our focus on developing foundation models that deliver exceptional utility and performance.
36BA3B256K ctx Apache 2.01.3M 1.6KFeb 2026- Same category
- Smaller (36B vs 219B)
- Commercial use OK
We introduce InternVL3, an advanced multimodal large language model (MLLM) series that demonstrates superior overall performance.
78.4B32K ctx Qwen License11.2K 239Apr 2025- Same category
- Smaller (78.4B vs 219B)
The Llama 4 collection of models are natively multimodal AI models that enable text and multimodal experiences. These models leverage a mixture-of-experts architecture to offer industry-leading performance in text and image understanding.
402BA17B256K ctx Llama 4 Community License206 105Apr 2025- Same category
- Similar size (402B vs 219B)
Hy4 is a new-generation Mixture-of-Experts (MoE) flagship model developed by the Tencent Hy Team, supporting text, image, and video understanding. The language backbone comprises 770B total parameters, of which 49B are activated per token.
774BA53.6B1M ctx tencent-hy-license974 174Oct 2026- Same category
- Larger (774B vs 219B)
Hy4 is a new-generation Mixture-of-Experts (MoE) flagship model developed by the Tencent Hy Team, supporting text, image, and video understanding. The language backbone comprises 770B total parameters, of which 49B are activated per token.
771BA50.3B1M ctx tencent-hy-license482 9Nov 2026- Same category
- Larger (771B vs 219B)
- Newer release
The main purpose of text image correction is to carry out geometric transformation on the image to correct the document distortion, inclination, perspective deformation and other problems in the image, so that the subsequent text recognition can be more accurate.
?BApache 2.0450.9K 14Jun 2025- Same category
- Commercial use OK
Following the February release of the Qwen3.5 series, we're pleased to share the first open-weight variant of Qwen3.6. Built on direct feedback from the community, Qwen3.6 prioritizes stability and real-world utility, offering developers a more intuitive, responsive, and genuinely productive coding…
27.8B256K ctx Apache 2.02.4M 2.4KApr 2026- Same category
- Smaller (27.8B vs 219B)
- Commercial use OK
- Widely used
💻Github Repo • 🤗Model Collections • 📜Technical Report • 💬Online Chat
917BA22.9B256K ctx Apache 2.011.6K 277Feb 2026- Same category
- Larger (917B vs 219B)
- Commercial use OK
Gemma is a family of open models built by Google DeepMind. Gemma 4 models are multimodal, handling text and image input (with audio supported on E2B, E4B, and 12B) and generating text output. This release includes open-weights models in both pre-trained and instruction-tuned variants.
25.8BA4B256K ctx Apache 2.04.8M 1.8KMar 2026- Same category
- Smaller (25.8B vs 219B)
- Commercial use OK
- Widely used
The Document Image Orientation Classification Module is primarily designed to distinguish the orientation of document images and correct them through post-processing.
?BApache 2.0326.3K 24Jun 2025- Same category
- Commercial use OK
This family of models performs vision-language and text-only tasks including optical character recognition, multimodal reasoning, localization, common sense reasoning, world knowledge utilization, and coding.
79.4B32K ctx CC BY-NC 4.07.9K 780Sept 2024- Same category
- Smaller (79.4B vs 219B)
We're excited to unveil Qwen2-VL, the latest iteration of our Qwen-VL model, representing nearly a year of innovation.
73.4B32K ctx Qwen License14.9K 318Sept 2024- Same category
- Smaller (73.4B vs 219B)
Over recent months, we have intensified our focus on developing foundation models that deliver exceptional utility and performance.
27.8B256K ctx Apache 2.01.6M 1.1KFeb 2026- Same category
- Smaller (27.8B vs 219B)
- Commercial use OK
Muse Glimmer is a 30-billion-parameter causal language model with a dedicated perception encoder, distilled from Muse Spark and purpose-built for autonomous agentic tasks on consumer hardware.
29.8B128K ctx Apache 2.0149.3K 2.2KAug 2026- Same category
- Smaller (29.8B vs 219B)
- Commercial use OK
UI-TARS is a next-generation native GUI agent model designed to interact seamlessly with graphical user interfaces (GUIs) using human-like perception, reasoning, and action capabilities.
73.4B32K ctx Apache 2.02.9K 159Jan 2025- Same category
- Smaller (73.4B vs 219B)
- Commercial use OK