Updated Daily FNI Ranked

AI Models Catalog 2,500

Open-weights intelligence — ranked by the Free2AITools Nexus Index (FNI).

Category:

Loading...

Sort:

image text to text

Mistral-Small-3.1-24B-Instruct-2503-FP8-dynamic

ARCH: Mistral3ForConditionalGeneration

🧠 24.01B Size

💾 ~20.5G VRAM

📜 Apache License

text generation

Qwen3.6-27B-MTPLX-Optimized-Speed

ARCH: Qwen3_5ForConditionalGeneration

🧠 4.67B Size

📜 Apache License

image text to text

NuExtract-2.0-8B-GPTQ

ARCH: Qwen2_5_VLForConditionalGeneration

🧠 8.29B Size

💾 ~8.7G VRAM

📜 MIT License

text generation

Qwen3-0.6B

ARCH: text generation

💾 ~1.8G VRAM

📜 Apache License

image text to text

Qwen3.6-27B-4bit

ARCH: Qwen3_5ForConditionalGeneration

🧠 4.67B Size

📜 Apache License

text generation

Llama-3.3-70B-Instruct-quantized.w8a8

ARCH: LlamaForCausalLM

🧠 70.56B Size

💾 ~55.4G VRAM

📜 llama3.3 License

gemma-4-19b-a4b-it-REAP-i1-GGUF

💾 ~15.6G VRAM

📜 Gemma License

CosyVoice2-0.5B

ARCH: text to speech

💾 ~1.7G VRAM

📜 Apache License

text generation

North-Mini-Code-1.0

ARCH: Cohere2MoeForCausalLM

🧠 30.48B Size

💾 ~25.4G VRAM

📜 Apache License

text generation

DeepSeek-V4-Flash-W4A16-FP8

ARCH: DeepseekV4ForCausalLM

🧠 44.1B Size

💾 ~35.6G VRAM

📜 MIT License

image text to text

Llama-4-Scout-17B-16E-Instruct-quantized.w4a16

ARCH: Llama4ForConditionalGeneration

🧠 109.44B Size

💾 ~84.6G VRAM

📜 llama4 License

text generation

Qwen3.6-35B-A3B-NVFP4

ARCH: Qwen3_5MoeForConditionalGeneration

🧠 18.68B Size

💾 ~16.5G VRAM

📜 Apache License

automatic speech recognition

SenseVoiceSmall

ARCH: automatic speech recognition

📜 Other License

text generation

Nemotron-Labs-Diffusion-8B

ARCH: NemotronLabsDiffusionModel

🧠 8.49B Size

💾 ~7.7G VRAM

📜 Other License

text generation

Gemma-4-21B

ARCH: Gemma4ForConditionalGeneration

🧠 21.34B Size

💾 ~17.3G VRAM

📜 Gemma License

text generation

Nemotron-Labs-Diffusion-3B

ARCH: NemotronLabsDiffusionModel

🧠 3.83B Size

💾 ~4.2G VRAM

📜 Other License

text generation

Ternary-Bonsai-1.7B-gguf

ARCH: text generation

💾 ~2.6G VRAM

📜 Apache License

image text to text

Step-3.7-Flash

ARCH: Step3p7ForConditionalGeneration

🧠 201.37B Size

💾 ~153.5G VRAM

📜 Apache License

text generation

GeM2-Llamion-14B-Base

ARCH: LlamaForCausalLM

🧠 14.5B Size

💾 ~12.2G VRAM

📜 Apache License

image text to text

typhoon-ocr-3b

ARCH: Qwen2_5_VLForConditionalGeneration

🧠 3.75B Size

💾 ~5.3G VRAM

📜 Apache License

text generation

Gemma-SEA-LION-v4-27B-IT

ARCH: Gemma3ForConditionalGeneration

🧠 27.43B Size

💾 ~21.9G VRAM

📜 Gemma License

LoRA_SDXL

ARCH: text to image

📜 Apache License

text generation

Ternary-Bonsai-4B-gguf

ARCH: text generation

💾 ~4.3G VRAM

📜 Apache License

text generation

GLM-5.1-478B-NVFP4

ARCH: GlmMoeDsaForCausalLM

🧠 280.22B Size

💾 ~212.7G VRAM

📜 MIT License

text generation

Trinity-Mini-Base

ARCH: AfmoeForCausalLM

🧠 26.12B Size

💾 ~20.9G VRAM

📜 Other License

text generation

Qwen-SEA-LION-v4-32B-IT

ARCH: Qwen3ForCausalLM

🧠 32.76B Size

💾 ~30.1G VRAM

📜 MIT License

NuExtract3

ARCH: Qwen3_5ForConditionalGeneration

🧠 4.54B Size

💾 ~5.9G VRAM

📜 Apache License

text generation

phi-4-quantized.w8a8

ARCH: Phi3ForCausalLM

🧠 14.66B Size

💾 ~12.3G VRAM

📜 MIT License

text generation

Qwen3.6-35B-GGUF

ARCH: text generation

💾 ~28.8G VRAM

📜 Apache License

text generation

Qwen3.6-28B-REAP

ARCH: Qwen3_5MoeForCausalLM

🧠 28.24B Size

💾 ~23.7G VRAM

📜 Apache License

image text to text

Ornstein3.6-35B-A3B-GGUF

ARCH: Qwen3_5MoeForConditionalGeneration

💾 ~31.8G VRAM

📜 Apache License

text generation

gpt-sw3-40b

ARCH: GPT2LMHeadModel

🧠 39.93B Size

💾 ~32.4G VRAM

📜 Other License

text generation

RavenX-CyberAgent-Qwen3.6-35B-A3B-Opus-4.7-OpenMythos-Pentester-BugHunter-RATH-mlx

ARCH: Qwen3_5MoeForConditionalGeneration

🧠 34.66B Size

💾 ~31.5G VRAM

📜 Apache License

text generation

GeM2-Llamion-14B-LongChat

ARCH: LlamaForCausalLM

🧠 14.5B Size

💾 ~12.2G VRAM

📜 Apache License

text generation

Qwen3.6-27B-MagicQuant-MTP-GGUF

ARCH: Qwen3_5ForConditionalGeneration

💾 ~22.8G VRAM

📜 Apache License

image text to text

Qwen3.5-9B-fp8

ARCH: Qwen3_5ForConditionalGeneration

🧠 9.65B Size

💾 ~9.7G VRAM

📜 Apache License

text generation

PLLuM-12B-nc-chat-2412

ARCH: MistralForCausalLM

🧠 12.25B Size

💾 ~11.7G VRAM

📜 CC-BY-NC-4.0 License

image text to text

Mistral-Small-3.1-24B-Instruct-2503-quantized.w4a16

ARCH: Mistral3ForConditionalGeneration

🧠 24.19B Size

💾 ~20.6G VRAM

📜 Apache License

text generation

PLLuM-12B-chat-2412

ARCH: MistralForCausalLM

🧠 12.25B Size

💾 ~11.7G VRAM

📜 Apache License

text generation

Llama-3.3-70B-Instruct-quantized.w4a16

ARCH: LlamaForCausalLM

🧠 71.14B Size

💾 ~55.9G VRAM

📜 llama3.3 License

text generation

archived-Kimi-K2.5-MLX-3.6bit

ARCH: text generation

text generation

gpt-sw3-1.3b

ARCH: GPT2LMHeadModel

🧠 1.44B Size

💾 ~2.4G VRAM

📜 Apache License

text generation

Racka-4B

ARCH: Qwen3ForCausalLM

🧠 4.02B Size

💾 ~5.5G VRAM

📜 CC-BY-NC-SA-4.0 License

UltraReal_FineTune_Anima_preview1_v1

ARCH: text to image

📜 Apache License

image text to text

Qwen-SEA-LION-v4-8B-VL

ARCH: Qwen3VLForConditionalGeneration

🧠 8.77B Size

💾 ~9.1G VRAM

📜 MIT License

text generation

gemma-2-9b-it-FP8

ARCH: Gemma2ForCausalLM

🧠 9.24B Size

💾 ~8.2G VRAM

📜 Gemma License

image text to text

Cosmos-Reason2-8B-NVFP4

ARCH: Qwen3VLForConditionalGeneration

🧠 5.73B Size

💾 ~6.8G VRAM

📜 Other License

text generation

Llama-3.1-8B-Instruct

ARCH: LlamaForCausalLM

🧠 8.03B Size

💾 ~7.3G VRAM

📜 llama3.1 License