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 851 to 900

Page 18 of 22

Model reference catalogue
Model TaskSizeContext LicenceEst. GPU Released
Aurora-Spec-Minimax-M2.1 togethercomputer · 1 build Aurora-Spec-Minimax-M2.1 Text gen858M 196,608 tokens Apache 2.0 ~2.5 GiB2026-02-04
MOSS-Audio-Tokenizer OpenMOSS-Team · 2 builds MOSS-Audio-Tokenizer Embeddings1.8B Apache 2.0 ~4.7 GiB2026-02-04
Aurora-Spec-Qwen3-Coder-Next togethercomputer · 1 build Aurora-Spec-Qwen3-Coder-Next-FP8 Text gen519M 262,144 tokens Apache 2.0 ~1.8 GiB2026-02-03
KD-Tinker Hugging Face · 1 build KD-Tinker Text gen8.2B 32,768 tokens ~18.6 GiB2026-02-03
LongCat-Image-Edit meituan-longcat · 1 build LongCat-Image-Edit-Turbo Image edit Apache 2.0 ~22.4 GiB2026-02-03
MiniCPM-o-4_5 openbmb · 4 builds MiniCPM-o-4_5 Omni (any→any)9.4B 40,960 tokens Apache 2.0 ~22.3 GiB2026-02-03
Sequential_Helium kyutai · 1 build Sequential_Helium_6B Text gen6.3B Open ~14.0 GiB2026-02-03
Intern-S1-Pro internlm · 2 builds Intern-S1-Pro Vision + text Apache 2.0 ~935.2 GiB2026-02-02
neodictabert-bilingual-embed dicta-il · 1 build neodictabert-bilingual-embed Embeddings363M Open ~1.8 GiB2026-02-02
SraVaani-1.0 ARTPARK-IISc · 1 build SraVaani-1.0 Speech→text MIT ~1.6 GiB2026-02-02
Qwen3-Coder-Next Qwen · 6 builds Qwen3-Coder-Next-FP8 Text gen79.7B 262,144 tokens Apache 2.0 ~82.9 GiB2026-02-01
Step-3.5-Flash stepfun-ai · 3 builds Step-3.5-Flash Text gen199.4B 262,144 tokens Apache 2.0 ~412.4 GiB2026-02-01
Step-3.5-Flash-Q4_K_S stepfun-ai · 1 build Step-3.5-Flash-GGUF-Q4_K_S Text gen 262,144 tokens Apache 2.0 ~121.5 GiB2026-02-01
GLM-OCR zai-org · 2 builds GLM-OCR Vision + text1.3B MIT ~5.1 GiB2026-01-30
acestep-continuous ACE-Step · 1 build acestep-v15-turbo-continuous Music / audio2.4B MIT ~6.5 GiB2026-01-29
acestep-shift1 ACE-Step · 1 build acestep-v15-turbo-shift1 Music / audio2.4B MIT ~6.5 GiB2026-01-29
MOVA-360p OpenMOSS-Team · 1 build MOVA-360p Image→video Apache 2.0 ~29.5 GiB2026-01-28
Olmo-Hybrid allenai · 1 build Olmo-Hybrid-7B Text gen7.4B 65,536 tokens Apache 2.0 ~16.8 GiB2026-01-28
PaddleOCR-VL-1.5 Baidu · 1 build PaddleOCR-VL-1.5 Vision + text959M 131,072 tokens Apache 2.0 ~4.0 GiB2026-01-28
Qwen3-ASR Qwen · 4 builds Qwen3-ASR-1.7B Speech→text2.3B Apache 2.0 ~6.4 GiB2026-01-28
Qwen3-ForcedAligner Qwen · 1 build Qwen3-ForcedAligner-0.6B Speech→text918M Apache 2.0 ~3.5 GiB2026-01-28
DeepSeek-OCR-2 DeepSeek · 1 build DeepSeek-OCR-2 Vision + text3.4B 8,192 tokens Apache 2.0 ~9.6 GiB2026-01-27
Trinity-Large arcee-ai · 4 builds Trinity-Large-Preview Text gen398.6B 262,144 tokens Licence fee ~810.7 GiB2026-01-27
Trinity-Large-TrueBase arcee-ai · 1 build Trinity-Large-TrueBase Text gen398.6B 8,192 tokens Licence fee ~810.7 GiB2026-01-27
C-RADIOv4-H NVIDIA · 1 build C-RADIOv4-H Embeddings652M Licence fee ~2.4 GiB2026-01-26
C-RADIOv4-SO400M NVIDIA · 1 build C-RADIOv4-SO400M Embeddings431M Licence fee ~2.0 GiB2026-01-26
EuroLLM-2512 utter-project · 2 builds EuroLLM-9B-Instruct-2512 Text gen9.2B 32,768 tokens Apache 2.0 ~20.7 GiB2026-01-26
LFM2.5-Audio-ONNX LiquidAI · 1 build LFM2.5-Audio-1.5B-ONNX Music / audio Licence fee 2026-01-26
mmbert-embed-32k-2d-matryoshka llm-semantic-router · 1 build mmbert-embed-32k-2d-matryoshka Embeddings307M 32,768 tokens Apache 2.0 ~1.7 GiB2026-01-26
SE-DiCoW BUT-FIT · 1 build SE-DiCoW Speech→text1.1B Apache 2.0 ~3.9 GiB2026-01-26
mmbert-embed-32k-2d-matryoshka vllm-sr · 1 build mmbert-embed-32k-2d-matryoshka Embeddings307M 32,768 tokens Apache 2.0 ~1.7 GiB2026-01-26
vllm-translategemma Infomaniak-AI · 3 builds vllm-translategemma-4b-it Vision + text5.0B 131,072 tokens Conditions ~11.7 GiB2026-01-26
HunyuanImage-3.0-Distil tencent · 1 build HunyuanImage-3.0-Instruct-Distil Image edit83.0B 22,800 tokens Licence fee ~11.0 GiB2026-01-25
Ace-Step1.5 ACE-Step · 1 build Ace-Step1.5 Music / audio MIT ~11.8 GiB2026-01-23
acestep ACE-Step · 1 build acestep-v15-base Music / audio2.4B MIT ~6.5 GiB2026-01-23
acestep-5Hz-lm ACE-Step · 2 builds acestep-5Hz-lm-4B Music / audio4.2B MIT ~10.1 GiB2026-01-23
acestep-captioner ACE-Step · 1 build acestep-captioner Music / audio10.7B MIT ~24.3 GiB2026-01-23
acestep-sft ACE-Step · 1 build acestep-v15-sft Music / audio2.4B MIT ~6.5 GiB2026-01-23
acestep-shift3 ACE-Step · 1 build acestep-v15-turbo-shift3 Music / audio2.4B MIT ~6.5 GiB2026-01-23
GLM-4.7-Flash-REAP cerebras · 1 build GLM-4.7-Flash-REAP-23B-A3B-GGUF Text gen · MoE MIT 2026-01-23
Phi-4-reasoning-vision microsoft · 1 build Phi-4-reasoning-vision-15B Vision + text15.1B 32,768 tokens MIT ~34.4 GiB2026-01-23
Youtu-VL tencent · 1 build Youtu-VL-4B-Instruct Vision + text5.3B 32,768 tokens Licence fee ~13.4 GiB2026-01-23
jina-embeddings-text-nano jinaai · 1 build jina-embeddings-v5-text-nano Embeddings212M 8,192 tokens Licence fee ~1.5 GiB2026-01-22
jina-embeddings-text-small jinaai · 1 build jina-embeddings-v5-text-small Embeddings596M 32,768 tokens Licence fee ~2.3 GiB2026-01-22
FLUX.2-klein kp-forks · 1 build FLUX.2-klein-9B Image edit9.1B Licence fee ~24.0 GiB2026-01-22
Qwen3-Embedding Qwen · 3 builds Qwen3-Embedding-4B Embeddings4.0B 40,960 tokens Apache 2.0 ~9.2 GiB2026-01-22
bge-m3 unsloth · 1 build bge-m3 Embeddings 8,194 tokens MIT ~3.4 GiB2026-01-22
powsm_ctc espnet · 1 build powsm_ctc Speech→text Open ~3.4 GiB2026-01-21
Qwen3-TTS-12Hz Qwen · 1 build Qwen3-TTS-12Hz-0.6B-Base Text→speech915M Apache 2.0 ~3.5 GiB2026-01-21
Qwen3-TTS-12Hz-CustomVoice Qwen · 2 builds Qwen3-TTS-12Hz-1.7B-CustomVoice Text→speech1.9B Apache 2.0 ~5.5 GiB2026-01-21

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