AxForge
Sign in Get started

Model catalogue

Explore models

Model families from the labs, their builds and hardware requirements.

Model type

Filters apply together. Press Enter in search or choose Apply filters.

Reset filters GPU catalogue · AxForge model offerings

1095 matching of 3295 lab model families · Showing 151 to 200

Page 4 of 22

Model reference catalogue
Model TaskSizeContext LicenceEst. GPU Released
NCP_ArchPreview_dolma3_8.9B_Stage1 ArchSpace-Collection · 1 build NCP_ArchPreview_dolma3_8.9B_Stage1 Text gen8.9B 8,192 tokens Apache 2.0 ~24.2 GiB2026-08-27
NCP_ArchPreview_dolma3_8.9B_Stage2 ArchSpace-Collection · 1 build NCP_ArchPreview_dolma3_8.9B_Stage2_v1 Text gen8.9B 8,192 tokens Apache 2.0 ~23.1 GiB2026-08-27
Qwen-Drive-1.0 Qwen · 1 build Qwen-Drive-1.0-4B Vision + text4.5B Apache 2.0 2026-08-27
SANA-Video_2.0_5B_720p_4step Efficient-Large-Model · 1 build SANA-Video_2.0_5B_720p_4step Text→video Apache 2.0 ~19.2 GiB2026-08-27
CASA-Helium1-VL-Shared kyutai · 1 build CASA-Helium1-VL-2B-Shared Vision + text2.7B 4,096 tokens Licence fee ~7.8 GiB2026-08-26
CASA-Qwen2_5-VL-Shared kyutai · 1 build CASA-Qwen2_5-VL-3B-Shared Vision + text3.8B 128,000 tokens Licence fee ~9.9 GiB2026-08-26
moss-tts-local-transformer-voice-acting-sft3 laion · 1 build moss-tts-local-transformer-4.55b-voice-acting-v2-sft3 Text→speech4.1B Open ~10.0 GiB2026-08-26
UI-Venus-2 inclusionAI · 1 build UI-Venus-2-9B Vision + text1M ~21.6 GiB2026-08-26
Breeze-TTS-2 BreezeBlue · 1 build Breeze-TTS-2 Text→speech3.5B Licence fee ~8.7 GiB2026-08-25
GLM-5.3 zai-org · 6 builds GLM-5.3 Text gen753.3B 1,048,576 tokens Licence fee ~768.0 GiB2026-08-25
GLM-5.3-Flash Z.ai · 7 builds GLM-5.3-Flash-NVFP4 Text genFlash 262,144 tokens MIT ~203.9 GiB2026-08-25
Moderato-Pro nitrai-research · 1 build Moderato-V1-Pro Text gen113.3B 262,144 tokens Apache 2.0 ~232.6 GiB2026-08-25
WeMM-Embedding tencent · 3 builds WeMM-Embedding-2B Embeddings2.7B Licence fee ~6.6 GiB2026-08-25
moonshine-streaming-tiny-es moonshine-ai · 1 build moonshine-streaming-tiny-es Speech→text27M 4,096 tokens MIT ~1.7 GiB2026-08-24
Qwen3.8-Flash-Next Qwen · 6 builds Qwen3.8-Flash-Next Vision + text180.0B Licence fee ~367.7 GiB2026-08-24
Spark-X2.5 XHToken · 3 builds Spark-X2.5-4B Text gen4.1B 1,048,576 tokens Apache 2.0 ~9.8 GiB2026-08-24
FastVideo-Minimax-FastH3 FastVideo · 1 build FastVideo-Minimax-FastH3-Preview-v0.2 Text→video35.0B Licence fee ~71.5 GiB2026-08-23
Muse2 Muse-research · 1 build Muse2-230M Text gen196M 128,000 tokens Apache 2.0 ~1.2 GiB2026-08-23
AtmicEmbeddingv3 SriRamanaAtmic · 1 build AtmicEmbeddingv3 Embeddings560M 514 tokens MIT ~2.2 GiB2026-08-22
AuK-Flash tencent · 1 build AuK-Flash Text→speech MIT ~8.5 GiB2026-08-21
gpt2-custom EleutherAI · 1 build gpt2-custom Text gen124M 1,024 tokens ~0.8 GiB2026-08-21
Supra2-Medium SupraLabs · 1 build Supra2-Medium-Instruct Text gen25M 1,024 tokens Apache 2.0 ~0.5 GiB2026-08-21
ArmorOCR inclusionAI · 2 builds ArmorOCR Vision + text8.8B Apache 2.0 ~21.0 GiB2026-08-20
Aurora-80K AuroraAI-Research · 1 build Aurora-80K Text gen Apache 2.0 ~0.4 GiB2026-08-20
SANA-Video_2.0_5B_720p Efficient-Large-Model · 1 build SANA-Video_2.0_5B_720p Text→video Apache 2.0 ~18.6 GiB2026-08-20
LFM2.5-DSpark LiquidAI · 1 build LFM2.5-2.6B-DSpark-GGUF Text gen Licence fee 2026-08-19
Qwen3.8-RTX5090-LMHead4 gittensor-model-hub · 1 build Qwen3.8-27B-NVFP4-RTX5090-LMHead4 Vision + text15.0B 262,144 tokens Apache 2.0 ~21.8 GiB2026-08-19
SenseNova-U1.5-MoT sensenova · 1 build SenseNova-U1.5-8B-MoT Omni (any→any)17.5B Apache 2.0 ~39.0 GiB2026-08-19
SenseNova-U1.5-MoT-SFT sensenova · 1 build SenseNova-U1.5-8B-MoT-SFT Omni (any→any)17.5B Apache 2.0 ~39.0 GiB2026-08-19
tipsv1-b14 Google · 1 build tipsv1-b14 Image embed196M Apache 2.0 ~1.5 GiB2026-08-19
tipsv1-g14 Google · 1 build tipsv1-g14 Image embed1.5B Apache 2.0 ~4.2 GiB2026-08-19
tipsv1-g14-lowres Google · 1 build tipsv1-g14-lowres Image embed1.5B Apache 2.0 ~4.2 GiB2026-08-19
tipsv1-l14 Google · 1 build tipsv1-l14 Image embed488M Apache 2.0 ~2.1 GiB2026-08-19
tipsv1-s14 Google · 1 build tipsv1-s14 Image embed56M Apache 2.0 ~1.2 GiB2026-08-19
tipsv1-so400m14 Google · 1 build tipsv1-so400m14 Image embed862M Apache 2.0 ~2.8 GiB2026-08-19
AuK tencent · 1 build AuK Text→speech MIT ~8.5 GiB2026-08-18
Ornith-1.5 ornith-ai · 8 builds Ornith-1.5-35B-A3B-NVFP4 Text gen · MoE19.5B 262,144 tokens MIT ~26.4 GiB2026-08-18
Apodex-1.1-mini apodex · 1 build Apodex-1.1-mini-NVFP4 Text gen19.8B Apache 2.0 ~27.2 GiB2026-08-17
granite-speech-5.0-turboctc-nc IBM · 1 build granite-speech-5.0-470m-turboctc-nc Speech→text473M Licence fee ~2.6 GiB2026-08-17
Bernini-R-Aux-FlashPack fal · 1 build Bernini-R-Aux-FlashPack Image→video Apache 2.0 2026-08-16
Puro thu-pacman · 1 build Puro-2B-Base Text gen2.0B 4,096 tokens Apache 2.0 ~5.3 GiB2026-08-16
Wan2.2-S2V FastVideo · 1 build Wan2.2-S2V-14B-Diffusers Image→video16.3B Apache 2.0 ~33.5 GiB2026-08-15
llm-jp-4-vl llm-jp · 1 build llm-jp-4-vl-9b Vision + text9.1B Apache 2.0 ~21.5 GiB2026-08-15
qwen-image-edit-2511-transformer-flashpack fal · 1 build qwen-image-edit-2511-bf16-transformer-flashpack Image edit Apache 2.0 ~22.7 GiB2026-08-15
ZeroTTS zeroweight-ai · 1 build ZeroTTS Text→speech MIT ~1.6 GiB2026-08-15
BananaMind-2-Pro BananaMind · 1 build BananaMind-2-Pro Text gen160M 3,072 tokens Licence fee ~0.9 GiB2026-08-14
MathForm openbmb · 1 build MathForm-8B Text gen8.2B 40,960 tokens Apache 2.0 ~18.6 GiB2026-08-14
NeoMME-Retriever-ST-dense Hcompany · 1 build NeoMME-260M-Retriever-ST-dense Embeddings263M 16,384 tokens Apache 2.0 ~1.6 GiB2026-08-14
NVIDIA-Nemotron-Labs-Teacher NVIDIA · 1 build NVIDIA-Nemotron-Labs-Teacher-Chat Text gen560.5B Licence fee 2026-08-14
NVIDIA-Nemotron-Labs-Teacher-Competition-Coding NVIDIA · 1 build NVIDIA-Nemotron-Labs-Teacher-Competition-Coding Text gen560.5B Licence fee 2026-08-14

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.

© 2026 AxForge · EU-hosted AI infrastructure Pricing Docs Trust Privacy Terms