type stringclasses 1 value | name stringlengths 4 99 |
|---|---|
model | nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-FP8 |
model | nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16 |
model | AIDC-AI/Ovis-Image-7B |
model | swiss-ai/Apertus-8B-2509 |
model | swiss-ai/Apertus-70B-2509 |
model | shisa-ai/shisa-v2.1-unphi4-14b |
model | shisa-ai/shisa-v2.1-qwen3-8b |
model | zerofata/MS3.2-PaintedFantasy-v3-24B |
model | zai-org/Glyph |
model | zai-org/AutoGLM-Phone-9B |
model | zai-org/AutoGLM-Phone-9B-Multilingual |
model | WeiboAI/VibeThinker-1.5B |
model | utter-project/EuroLLM-22B-Instruct-2512 |
model | stepfun-ai/GELab-Zero-4B-preview |
model | squ11z1/Hypnos-i1-8B |
model | SonarSource/SonarSweep-java-gpt-oss-20b |
model | ServiceNow/GroundNext-7B-V0 |
model | SamuelBang/AesCoder-4B |
model | p-e-w/Qwen3-4B-Instruct-2507-heretic |
model | open-thoughts/OpenThinker-Agent-v1 |
model | nn-tech/MetalGPT-1 |
model | NiuTrans/LMT-60-8B |
model | mlfoundations/Gelato-30B-A3B |
model | miromind-ai/MiroThinker-v1.5-30B |
model | LiquidAI/LFM2-8B-A1B |
model | janhq/Jan-v2-VL-med |
model | janhq/Jan-v2-VL-low |
model | janhq/Jan-v2-VL-high |
model | internlm/JanusCoderV-8B |
model | internlm/JanusCoderV-7B |
model | internlm/JanusCoder-8B |
model | internlm/JanusCoder-14B |
model | Hcompany/Holo2-8B |
model | Hcompany/Holo2-4B |
model | bartowski/aquif-ai_aquif-3.5-Max-1205-GGUF |
model | AvitoTech/avibe |
model | allenai/Olmo-3.1-7B-RL-Zero-Code |
model | ai-sage/GigaChat3-10B-A1.8B |
model | ibm-granite/granite-4.0-1b-speech |
model | microsoft/VibeVoice-ASR-HF |
model | Qwen/Qwen3.5-397B-A17B-GPTQ-Int4 |
model | Qwen/Qwen3.5-35B-A3B-GPTQ-Int4 |
model | Qwen/Qwen3.5-27B-GPTQ-Int4 |
model | Qwen/Qwen3.5-122B-A10B-GPTQ-Int4 |
model | ibm-granite/granite-guardian-3.2-8b-factuality-detection |
model | Qwen/Qwen3.5-9B |
model | Qwen/Qwen3.5-4B |
model | Qwen/Qwen3.5-2B |
model | Qwen/Qwen3.5-0.8B |
model | marksverdhei/GLM-4.7-Flash-FP8 |
model | cerebras/GLM-4.6-REAP-218B-A32B-FP8 |
model | zai-org/GLM-4.6-FP8 |
model | zai-org/GLM-4.1V-9B-Thinking |
model | RUC-DataLab/DeepAnalyze-8B |
model | OpenGVLab/ScaleCUA-32B |
model | nvidia/NVIDIA-Nemotron-Nano-9B-v2 |
model | nvidia/NVIDIA-Nemotron-Nano-12B-v2 |
model | lm-provers/QED-Nano |
model | LiquidAI/LFM2-350M-PII-Extract-JP |
model | LiquidAI/LFM2-350M-Extract |
model | LiquidAI/LFM2-1.2B-Tool |
model | LiquidAI/LFM2-1.2B-Extract |
model | inclusionAI/ZwZ-4B |
model | Hcompany/Holo1.5-7B |
model | Hcompany/Holo1.5-3B |
model | foreverlasting1202/QuestA-Nemotron-1.5B |
model | baidu/ERNIE-4.5-21B-A3B-PT |
model | baidu/ERNIE-4.5-21B-A3B-Thinking |
model | baidu/ERNIE-4.5-0.3B-PT |
model | Qwen/Qwen3-4B |
model | Qwen/Qwen3-1.7B |
model | OPPOer/Qwen-Image-Pruning |
model | nvidia/Cosmos-Reason1-7B |
model | numind/NuMarkdown-8B-Thinking |
model | LiquidAI/LFM2-350M-Math |
model | LiquidAI/LFM2-1.2B |
model | kurakurai/Luth-LFM2-700M |
model | kurakurai/Luth-LFM2-350M |
model | kurakurai/Luth-LFM2-1.2B |
model | Qwen/Qwen3.5-35B-A3B-FP8 |
model | Qwen/Qwen3.5-27B-FP8 |
model | Qwen/Qwen3.5-122B-A10B-FP8 |
model | bharatgenai/Param-1-2.9B-Instruct |
model | bharatgenai/LegalParam |
model | bharatgenai/FinanceParam |
model | bharatgenai/AyurParam |
model | bharatgenai/AgriParam |
model | knowledge-computing/geolm-base-toponym-recognition |
model | Qwen/Qwen3.5-35B-A3B |
model | Qwen/Qwen3.5-35B-A3B-Base |
model | Qwen/Qwen3.5-27B |
model | Qwen/Qwen3.5-122B-A10B |
model | is36e/detr-resnet-50-sku110k |
model | is36e/detr-resnet-101-dc5-sku110k |
model | MiniMaxAI/MiniMax-M2.5 |
model | BSC-LT/ALIA-40b-instruct-2601 |
model | EleutherAI/pythia-31m-deduped |
model | EleutherAI/pythia-14m-deduped |
model | Qwen/Qwen3-32B |
model | openai/whisper-large-v3-turbo |
End of preview. Expand
in Data Studio
README.md exists but content is empty.
- Downloads last month
- 112