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 901 to 950

Page 19 of 22

Model reference catalogue
Model TaskSizeContext LicenceEst. GPU Released
Qwen3-TTS-12Hz-VoiceDesign Qwen · 1 build Qwen3-TTS-12Hz-1.7B-VoiceDesign Text→speech1.9B Apache 2.0 ~5.5 GiB2026-01-21
Qwen3-TTS-Tokenizer-12Hz Qwen · 1 build Qwen3-TTS-Tokenizer-12Hz Audio→audio171M Apache 2.0 ~2.0 GiB2026-01-21
VibeVoice-ASR microsoft · 1 build VibeVoice-ASR Speech→text8.7B MIT ~19.2 GiB2026-01-21
Voxtral-Mini-Realtime-2602 Mistral AI · 1 build Voxtral-Mini-4B-Realtime-2602 Speech→text4.4B Apache 2.0 ~10.6 GiB2026-01-21
INTELLECT-3.1 PrimeIntellect · 1 build INTELLECT-3.1 Text gen106.9B 131,072 tokens MIT ~219.6 GiB2026-01-20
pplx-embed-context perplexity-ai · 2 builds pplx-embed-context-v1-4b Embeddings4.0B 32,768 tokens MIT ~9.2 GiB2026-01-20
BFL-FLUX.2-klein Runware · 2 builds BFL-FLUX.2-klein-4B Image edit3.9B ~13.6 GiB2026-01-19
GLM-4.7-Flash zai-org · 3 builds GLM-4.7-Flash Text gen31.2B 202,752 tokens MIT ~64.7 GiB2026-01-19
MolDeBERTa-contrastive_mlc SaeedLab · 1 build MolDeBERTa-base-123M-contrastive_mlc Embeddings88M 128 tokens Licence fee ~1.3 GiB2026-01-19
MolDeBERTa-contrastive_mtr SaeedLab · 1 build MolDeBERTa-base-123M-contrastive_mtr Embeddings88M 128 tokens Licence fee ~1.3 GiB2026-01-19
MolDeBERTa-mlc SaeedLab · 1 build MolDeBERTa-base-123M-mlc Embeddings88M 128 tokens Licence fee ~1.3 GiB2026-01-19
MolDeBERTa-mlm SaeedLab · 1 build MolDeBERTa-base-123M-mlm Embeddings89M 128 tokens Licence fee ~1.3 GiB2026-01-19
MolDeBERTa-mtr SaeedLab · 1 build MolDeBERTa-base-123M-mtr Embeddings88M 128 tokens Licence fee ~1.3 GiB2026-01-19
MolDeBERTa-tiny-mlm SaeedLab · 1 build MolDeBERTa-tiny-10M-mlm Embeddings7M 128 tokens Licence fee ~1.1 GiB2026-01-19
SkyReels-A2V Skywork · 1 build SkyReels-V3-A2V-19B Image→video18.9B Licence fee ~38.8 GiB2026-01-19
SkyReels-R2V Skywork · 1 build SkyReels-V3-R2V-14B Image→video14.3B Licence fee ~29.5 GiB2026-01-19
SkyReels-V2V Skywork · 1 build SkyReels-V3-V2V-14B Video edit Licence fee ~29.5 GiB2026-01-19
LightOnOCR-2 lightonai · 1 build LightOnOCR-2-1B Vision + text1.0B Apache 2.0 ~4.9 GiB2026-01-16
LightOnOCR-2-bbox lightonai · 1 build LightOnOCR-2-1B-bbox Vision + text1.0B Apache 2.0 ~4.9 GiB2026-01-16
LightOnOCR-2-bbox-soup lightonai · 1 build LightOnOCR-2-1B-bbox-soup Vision + text1.0B Apache 2.0 ~4.9 GiB2026-01-16
LightOnOCR-2-ocr-soup lightonai · 1 build LightOnOCR-2-1B-ocr-soup Vision + text1.0B 16,384 tokens Apache 2.0 ~4.9 GiB2026-01-16
Mursit-Large-TR-Retrieval newmindai · 1 build Mursit-Large-TR-Retrieval Embeddings404M 2,048 tokens Apache 2.0 ~1.9 GiB2026-01-16
Mursit-TR-Retrieval newmindai · 1 build Mursit-Base-TR-Retrieval Embeddings156M 1,024 tokens Apache 2.0 ~1.4 GiB2026-01-16
PP-OCRv5_server_det_safetensors Baidu · 1 build PP-OCRv5_server_det_safetensors Image→text22M Apache 2.0 2026-01-16
parakeet-ctc-Vietnamese NVIDIA · 1 build parakeet-ctc-0.6b-Vietnamese Speech→text Licence fee ~1.8 GiB2026-01-15
PP-OCRv5_mobile_det_safetensors Baidu · 1 build PP-OCRv5_mobile_det_safetensors Image→text4M Apache 2.0 2026-01-15
Stable-DiffCoder ByteDance · 1 build Stable-DiffCoder-8B-Instruct Text gen8.3B MIT ~18.6 GiB2026-01-15
FLUX.2 Klein Black Forest Labs · 7 builds FLUX.2-klein-4B Image edit4B Apache 2.0 ~13.7 GiB2026-01-14
japanese-avhubert-large_noise_pt enactic · 1 build japanese-avhubert-large_noise_pt Embeddings325M Licence fee ~1.7 GiB2026-01-14
pplx-embed perplexity-ai · 2 builds pplx-embed-v1-0.6b Embeddings596M 32,768 tokens MIT ~2.3 GiB2026-01-14
Baichuan-M3 baichuan-inc · 3 builds Baichuan-M3-235B-GPTQ-INT4 Text gen235.1B 40,960 tokens Apache 2.0 ~129.4 GiB2026-01-13
Falcon-H1-Tiny-Coder tiiuae · 1 build Falcon-H1-Tiny-Coder-90M Text gen91M 262,144 tokens Licence fee ~0.8 GiB2026-01-13
Step3-VL stepfun-ai · 2 builds Step3-VL-10B Vision + text10.2B Apache 2.0 ~23.9 GiB2026-01-13
Unipic3 Skywork · 1 build Unipic3 Omni (any→any) MIT 2026-01-13
Falcon-H1-Tiny tiiuae · 1 build Falcon-H1-Tiny-90M-Instruct Text gen91M 262,144 tokens Licence fee ~0.8 GiB2026-01-12
Falcon-H1-Tiny-Multilingual tiiuae · 1 build Falcon-H1-Tiny-Multilingual-100M-Instruct Text gen108M 262,144 tokens Licence fee ~0.8 GiB2026-01-12
Falcon-H1-Tiny-R tiiuae · 2 builds Falcon-H1-Tiny-R-0.6B Text gen622M 262,144 tokens Licence fee ~2.0 GiB2026-01-12
Falcon-H1-Tiny-R-pre-GRPO tiiuae · 1 build Falcon-H1-Tiny-R-0.6B-pre-GRPO Text gen622M 262,144 tokens Licence fee ~2.0 GiB2026-01-12
speaker-detection aTrain-core · 1 build speaker-detection Speech→text Open ~1.7 GiB2026-01-12
translategemma Google · 3 builds translategemma-4b-it Vision + text5.0B Conditions 2026-01-12
Unipic3-Consistency-Model Skywork · 1 build Unipic3-Consistency-Model Omni (any→any) MIT 2026-01-12
Unipic3-DMD Skywork · 1 build Unipic3-DMD Omni (any→any) MIT 2026-01-12
kugelaudio-0-open kugelaudio · 1 build kugelaudio-0-open Text→speech9.3B MIT ~20.5 GiB2026-01-11
GLM-Image zai-org · 1 build GLM-Image Text→image6.9B MIT ~26.1 GiB2026-01-08
paza-whisper-large microsoft · 1 build paza-whisper-large-v3-turbo Speech→text809M MIT ~3.3 GiB2026-01-08
siglip2_decoder nyu-visionx · 1 build siglip2_decoder Image edit MIT ~7.3 GiB2026-01-08
SoulX-FlashTalk Soul-AILab · 1 build SoulX-FlashTalk-14B Image→video18.9B Apache 2.0 ~38.8 GiB2026-01-08
wav2vec2-french-phonemizer Cnam-LMSSC · 1 build wav2vec2-french-phonemizer-v2 Speech→text94M MIT ~1.8 GiB2026-01-08
medgemma-1.5 Google · 2 builds medgemma-1.5-4b-it Vision + text4.3B Licence fee 2026-01-07
Qwen3-VL-Embedding Qwen · 2 builds Qwen3-VL-Embedding-2B Embeddings2.1B Apache 2.0 ~5.4 GiB2026-01-07

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