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…
Alternatives to Mistral-Large-4-1T-A52B
120 open-weight models that do the same job as Mistral-Large-4-1T-A52B (1T 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.
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
- Smaller (402B vs 1T)
Over recent months, we have intensified our focus on developing foundation models that deliver exceptional utility and performance.
125BA10B256K ctx Apache 2.0672.5K 659Feb 2026- Same category
- Smaller (125B vs 1T)
- Commercial use OK
Over recent months, we have intensified our focus on developing foundation models that deliver exceptional utility and performance.
2.3B256K ctx Apache 2.05.4M 498Feb 2026- Same category
- Smaller (2.3B vs 1T)
- Commercial use OK
- Widely used
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…
36BA3B256K ctx Apache 2.03.8M 3.2KApr 2026- Same category
- Smaller (36B vs 1T)
- Commercial use OK
- Widely used
GitHub · On-device results · Evaluation · Examples · Usage · Model weights
35.2BA3B256K ctx Apache 2.0875.7K 1.1KOct 2026- Same category
- Smaller (35.2B vs 1T)
- Commercial use OK
- Newer release
Over recent months, we have intensified our focus on developing foundation models that deliver exceptional utility and performance.
4.7B256K ctx Apache 2.04.3M 1.3KFeb 2026- Same category
- Smaller (4.7B vs 1T)
- Commercial use OK
- Widely used
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 1T)
- Commercial use OK
- Widely used
NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16 is the full-precision release of NVIDIA Nemotron 3.5 Super VL, a vision-language model for reasoning, coding, tool use, document understanding, and video understanding.
124BA12B1M ctx openmdw-1.181K 136Oct 2026- Same category
- Smaller (124B vs 1T)
- Commercial use OK
- Newer release
MiniCPM-V 4.7 is the latest model in the MiniCPM-V series, available in a 1B variant. It builds on the family's single-image, multi-image, and video understanding capabilities, combining stronger foundation capabilities with efficient visual encoding and on-device inference.
1.6B256K ctx Apache 2.0468.8K 418Oct 2026- Same category
- Smaller (1.6B vs 1T)
- Commercial use OK
- Newer release
Over recent months, we have intensified our focus on developing foundation models that deliver exceptional utility and performance.
873M256K ctx Apache 2.02.5M 856Feb 2026- Same category
- Smaller (873M vs 1T)
- Commercial use OK
- Widely used
Agnes 2.5 Pro Theta is a multimodal mixture-of-experts checkpoint distributed with its model implementation, tokenizer, chat template, and image/video processors. This release provides BF16 weights for local inference through Transformers.
188BApache 2.01.1K 54Nov 2026- Same category
- Smaller (188B vs 1T)
- Commercial use OK
- Newer release
Meet Qwen3-VL — the most powerful vision-language model in the Qwen series to date.
8.8B256K ctx Apache 2.06.1M 1.3KOct 2025- Same category
- Smaller (8.8B vs 1T)
- Commercial use OK
- Widely used
NVIDIA Nemotron Parse 2.0 transforms document images into structured, machine-readable representations with text, layout classes, bounding boxes, and reading-order information.
903M9K ctx Custom licence650.6K 136Jun 2026- Same category
- Smaller (903M vs 1T)
- Commercial use OK
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 1T)
- Commercial use OK
DiffusionGemma is a generative model built by Google DeepMind. Based on the 26B A4B Mixture-of-Experts (MoE) Gemma 4 architecture, DiffusionGemma generates tokens using discrete diffusion. This open-weights model is multimodal, handling text, image, and video inputs to generate text output.
25.8BA4B256K ctx Apache 2.0563.7K 1.3KJun 2026- Same category
- Smaller (25.8B vs 1T)
- Commercial use OK
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 1T)
- Commercial use OK
💻 Run it on your own machine: POCKET-Darwin-180B-GGUF, the 4-bit GGUF of R3 (111 GB), runs on a laptop with an 8 GB GPU and 32 GB RAM, CPU only at 18–21 tok/s, a 128 GB mini PC or one DGX Spark. MMLU-Pro is identical to BF16 (87.65%).
180BA61.6B256K ctx qwen-community-1.01.1K 49Oct 2026- Same category
- Smaller (180B vs 1T)
- Newer release
Meet Qwen3-VL — the most powerful vision-language model in the Qwen series to date.
4.4B256K ctx Apache 2.02.8M 544Oct 2025- Same category
- Smaller (4.4B vs 1T)
- Commercial use OK
- Widely used
Over recent months, we have intensified our focus on developing foundation models that deliver exceptional utility and performance.
873M256K ctx Apache 2.0880.7K 142Feb 2026- Same category
- Smaller (873M vs 1T)
- Commercial use OK
💻 Run it on your own machine — POCKET-Darwin-180B-GGUF: the 4-bit GGUF of R3 (111 GB) runs on a laptop with an 8 GB GPU and 32 GB RAM, CPU only at 18–21 tok/s, a 128 GB mini PC or one DGX Spark — MMLU-Pro identical to BF16 (87.65%).
180BA61.6B256K ctx qwen-community-1.0759 140Sept 2026- Same category
- Smaller (180B vs 1T)
This is an uncensored version of Qwen/Qwen3.8-27B created with abliteration (see remove-refusals-with-transformers to know more about it). This is a crude, proof-of-concept implementation to remove refusals from an LLM model without using TransformerLens.
27.8B256K ctx Apache 2.0170.2K 819Aug 2026- Same category
- Smaller (27.8B vs 1T)
- 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 1T)
- Commercial use OK
Over recent months, we have intensified our focus on developing foundation models that deliver exceptional utility and performance.
4.7B256K ctx Apache 2.0496.7K 133Feb 2026- Same category
- Smaller (4.7B vs 1T)
- Commercial use OK
ZAYA1-VL-8B is a vision-language model (VLM) built upon Zyphra's ZAYA1-8B LLM. It has state-of-the-art performance among VLMs for its size and inference efficiency.
9.7B32K ctx Apache 2.0339.4K 43Apr 2026- Same category
- Smaller (9.7B vs 1T)
- Commercial use OK