Model catalogue
1095 matching of 3295 lab model families · Showing 401 to 450
Page 9 of 22
| Model | Task | Size | Context | Licence | Est. GPU | Released |
|---|---|---|---|---|---|---|
| lift datalab-to · 1 build lift | Vision + text | 9.7B | 262,144 tokens | Open | ~22.1 GiB | 2026-06-19 |
| react-native-executorch-lfm2.5-embedding software-mansion · 1 build react-native-executorch-lfm2.5-embedding-350m | Embeddings | Licence fee | 2026-06-19 | |||
| Unlimited-OCR Baidu · 1 build Unlimited-OCR | Vision + text | 3.3B | 32,768 tokens | MIT | ~9.5 GiB | 2026-06-19 |
| cohere-transcribe-arabic-07-2026 Cohere · 1 build cohere-transcribe-arabic-07-2026 | Speech→text | 2.1B | Apache 2.0 | ~5.8 GiB | 2026-06-18 | |
| Krea-2 krea · 1 build Krea-2-Turbo | Text→image | 12.8B | Licence fee | ~32.2 GiB | 2026-06-18 | |
| Krea-2-Raw krea · 2 builds Krea-2-Raw | Text→image | 12.8B | Licence fee | ~32.2 GiB | 2026-06-18 | |
| Echo-DSRN.3-Embed-Exp ethicalabs · 1 build Echo-DSRN-v0.1.3-Embed-Exp | Embeddings | 98M | 2,048 tokens | ~1.3 GiB | 2026-06-17 | |
| Boogu-Image-0.1-Edit Boogu · 1 build Boogu-Image-0.1-Edit | Image edit | 10.3B | Apache 2.0 | ~26.4 GiB | 2026-06-16 | |
| eegdino-medium-pretrained braindecode · 1 build eegdino-medium-pretrained | Embeddings | 34M | BSD | ~1.2 GiB | 2026-06-16 | |
| eegdino-small-pretrained braindecode · 1 build eegdino-small-pretrained | Embeddings | 5M | BSD | ~1.1 GiB | 2026-06-16 | |
| FastWan-QAD FastVideo · 2 builds FastWan-QAD-1.3B | Text→video | 1.4B | Apache 2.0 | ~12.0 GiB | 2026-06-16 | |
| GLM-5.2 zai-org · 4 builds GLM-5.2 | Text gen | 753.3B | 1,048,576 tokens | MIT | ~1528.3 GiB | 2026-06-16 |
| UniAR-RL ShareLab-SII · 1 build UniAR-RL | Image→text | 9.6B | 32,768 tokens | Apache 2.0 | ~22.8 GiB | 2026-06-16 |
| UniAR-SFT ShareLab-SII · 1 build UniAR-SFT | Image→text | 9.6B | 32,768 tokens | Apache 2.0 | ~22.8 GiB | 2026-06-16 |
| LFM2.5-ONNX LiquidAI · 1 build LFM2.5-230M-ONNX | Text gen | 128,000 tokens | Licence fee | 2026-06-15 | ||
| MolmoMotion-H1-F32 allenai · 1 build MolmoMotion-4B-H1-F32 | Vision + text | 4.9B | Apache 2.0 | ~11.5 GiB | 2026-06-15 | |
| MolmoMotion-H3-F30 allenai · 1 build MolmoMotion-4B-H3-F30 | Vision + text | 4.9B | Apache 2.0 | ~11.5 GiB | 2026-06-15 | |
| Sa2VA-LLaVA-1.5 ByteDance · 1 build Sa2VA-LLaVA-1.5-7B | Vision + text | 7.3B | Apache 2.0 | ~18.9 GiB | 2026-06-15 | |
| voiceclap-commercial laion · 1 build voiceclap-commercial | Embeddings | 113M | Open | ~1.3 GiB | 2026-06-15 | |
| Kimi-K3 moonshotai · 3 builds Kimi-K3 | Vision + text | 2779.9B | Licence fee | 2026-06-13 | ||
| Kimi-K2.7-Code moonshotai · 3 builds Kimi-K2.7-Code | Vision + text | 1026.9B | Licence fee | ~606.8 GiB | 2026-06-11 | |
| MobileMoE-L-QAT Meta · 1 build MobileMoE-L-QAT | Text gen · MoE | 2.8B | Licence fee | 2026-06-11 | ||
| MobileMoE-L-SFT Meta · 1 build MobileMoE-L-SFT | Text gen · MoE | 5.3B | Licence fee | 2026-06-11 | ||
| MobileMoE-M Meta · 1 build MobileMoE-M-Base | Text gen · MoE | 2.8B | Licence fee | 2026-06-11 | ||
| MobileMoE-M-QAT Meta · 1 build MobileMoE-M-QAT | Text gen · MoE | 1.5B | Licence fee | 2026-06-11 | ||
| MobileMoE-M-SFT Meta · 1 build MobileMoE-M-SFT | Text gen · MoE | 2.8B | Licence fee | 2026-06-11 | ||
| MobileMoE-S Meta · 1 build MobileMoE-S-Base | Text gen · MoE | 1.3B | Licence fee | 2026-06-11 | ||
| MobileMoE-S-QAT Meta · 1 build MobileMoE-S-QAT | Text gen · MoE | 672M | Licence fee | 2026-06-11 | ||
| MobileMoE-S-SFT Meta · 1 build MobileMoE-S-SFT | Text gen · MoE | 1.3B | Licence fee | 2026-06-11 | ||
| PP-OCRv6_tiny_rec Baidu · 1 build PP-OCRv6_tiny_rec | Image→text | Apache 2.0 | 2026-06-11 | |||
| prxpixel-t2i Photoroom · 1 build prxpixel-t2i | Text→image | 7.0B | Apache 2.0 | ~19.8 GiB | 2026-06-11 | |
| voiceclap-small laion · 1 build voiceclap-small-v2 | Embeddings | 113M | Licence fee | ~1.3 GiB | 2026-06-11 | |
| ZONOS2 Zyphra · 1 build ZONOS2 | Text→speech | Apache 2.0 | ~17.2 GiB | 2026-06-11 | ||
| EvoQuality ByteDance · 1 build EvoQuality | Vision + text | 8.3B | Apache 2.0 | ~19.4 GiB | 2026-06-10 | |
| KDL-Frontier-Parser-nano KDLAI · 1 build KDL-Frontier-Parser-nano | Vision + text | 1.2B | 8,192 tokens | Licence fee | ~4.4 GiB | 2026-06-10 |
| MobileMoE-L Meta · 1 build MobileMoE-L-Base | Text gen · MoE | 5.3B | Licence fee | 2026-06-10 | ||
| parakeet-ctc-farsi Peacockery · 1 build parakeet-ctc-109m-farsi | Speech→text | Open | 2026-06-10 | |||
| PP-OCRv6_medium_det Baidu · 1 build PP-OCRv6_medium_det | Image→text | Apache 2.0 | 2026-06-10 | |||
| PP-OCRv6_medium_rec Baidu · 1 build PP-OCRv6_medium_rec | Image→text | Apache 2.0 | 2026-06-10 | |||
| PP-OCRv6_small_det Baidu · 1 build PP-OCRv6_small_det | Image→text | Apache 2.0 | 2026-06-10 | |||
| PP-OCRv6_small_rec Baidu · 1 build PP-OCRv6_small_rec | Image→text | Apache 2.0 | 2026-06-10 | |||
| PP-OCRv6_tiny_det Baidu · 1 build PP-OCRv6_tiny_det | Image→text | Apache 2.0 | 2026-06-10 | |||
| Sa2VA-Qwen3-VL-SAM3 ByteDance · 1 build Sa2VA-Qwen3-VL-4B-SAM3 | Vision + text | 5.3B | Apache 2.0 | ~14.0 GiB | 2026-06-10 | |
| SANA-WM_chunk_causal Efficient-Large-Model · 1 build SANA-WM_chunk_causal | Text→video | Apache 2.0 | ~11.3 GiB | 2026-06-10 | ||
| Supra1.5-exp SupraLabs · 1 build Supra1.5-50M-Base-exp | Text gen | 52M | 5,120 tokens | Apache 2.0 | ~0.7 GiB | 2026-06-10 |
| ta_openslr127 espnet · 1 build ta_openslr127 | Speech→text | Open | ~1.8 GiB | 2026-06-10 | ||
| UltraX openbmb · 1 build UltraX-0.6B-Preview | Text gen | Apache 2.0 | 2026-06-10 | |||
| diffusiongemma Google · 5 builds diffusiongemma-26B-A4B-it | Vision + text · MoE | 25.8B | Apache 2.0 | ~55.8 GiB | 2026-06-09 | |
| GLACIER-100k-MiniMol glacier-hf · 1 build GLACIER-100k-MiniMol | Embeddings | MIT | ~1.1 GiB | 2026-06-09 | ||
| PP-OCRv6_medium_det_onnx Baidu · 1 build PP-OCRv6_medium_det_onnx | Image→text | Apache 2.0 | 2026-06-09 |
Release dates use the upstream repository's publication date. The default is the past 12 months, through today. All time includes unknown and future dates. 3 families have no recorded date.
Size, context and GPU memory describe the listed build. For language models, the memory estimate covers one request at 8K context (or the model's shorter limit); image estimates cover one image. Other types use their existing weight and overhead estimates. Blank values are missing or incomplete and do not match numeric limits. Open a model to compare its builds and hardware estimates.