Model catalog

All models tracked for private deployments.

Each model is evaluated by modality, memory footprint, status and fit with local infrastructure constraints.

Back to integration

The 3 latest validated models

HF Trending

GLM-5.2

Z AI · 2026-06-17

Modality: text → text
Parameters: 753B total · active not specified
Size: Not specified

HF Trending

Nemotron 3.5 ASR Streaming

Nvidia · 2026-06-16

Modality: audio → text
Parameters: 0.6B total · active not specified
Size: Not specified

HF Trending

MiniMax-M3

MiniMax AI · 2026-06-15

Modality: image, text → text
Parameters: 427B total · active not specified
Size: Not specified

The 3 best-known models

Testing

Gemma 4

Google · 2026-04-02

Modality: text, image, video → text, tool, code
Parameters: 31B total · 31B active
Size: 32gb

Stable

DeepSeek-V4-Pro

DeepSeek AI · 2026-06-07

Modality: text → text
Parameters: 862B total · active not specified
Size: Not specified

Testing

Qwen 3.6 VL

Alibaba Cloud · 2026-04-24

Modality: image, text → text
Parameters: 35B total · 3B active
Size: 24gb

Stable

Kimi-K2.7-Code

Moonshot AI · 2026-06-14

Modality: image, text → text, code
Parameters: 1.1T total · active not specified
Size: Not specified

Stable

diffusiongemma-26B-A4B-it

Google · 2026-06-10

Modality: image, text → text
Parameters: 26B total · 4B active
Size: Not specified

Stable

LocateAnything-3B

Nvidia · 2026-06-12

Modality: image, text → text
Parameters: 4B total · active not specified
Size: Not specified

HF Trending

North-Mini-Code-1.0

Cohere Labs · 2026-06-14

Modality: text, code → text, code
Parameters: 30B total · active not specified
Size: Not specified

Stable

Nex-N2-Pro

Nex AGI · 2026-06-11

Modality: text → text
Parameters: 397B total · active not specified
Size: Not specified

Stable

SCAIL-2

Z AI · 2026-06-15

Modality: image → video
Parameters: Not specified
Size: Not specified

Testing

Nemotron OCR v2

Nvidia · 2026-04-02

Modality: ocr → text
Parameters: 0.1B total · 0.1B active
Size: 0.4gb

Testing

Cohere Transcribe

Cohere Labs · 2026-03-25

Modality: audio → text
Parameters: 2B total · 2B active
Size: 2gb

Stable

Nemotron 3 Super

Nvidia · 2026-03-10

Modality: text, code → text, tool, code
Parameters: 124B total · 12B active
Size: 74gb

Stable

LTX-2.3

Lightricks · 2026-03-03

Modality: text → video
Parameters: 22B total · 22B active
Size: 20gb

Stable

Qwen 3.5

Alibaba Cloud · 2026-02-16

Modality: text, image → text, tool
Parameters: 397B total · 17B active
Size: 233gb

Stable

MiniMax M2.5

MiniMax AI · 2026-02-12

Modality: text, code → text, tool, code
Parameters: 229B total · 10B active
Size: 130gb

Stable

Step 3.5 Flash

Stepfun AI · 2026-02-11

Modality: code → code, tool
Parameters: 199B total · 11B active
Size: 194gb

Stable

GLM 5

Z AI · 2026-02-10

Modality: text, code → text, tool, code
Parameters: 435B total · 40B active
Size: 429gb

Stable

Qwen 3 Coder Next

Alibaba Cloud · 2026-02-03

Modality: code → code, tool
Parameters: 80B total · 3B active
Size: 45gb

Stable

Paddle OCR VL 1.5

Baidu · 2026-01-28

Modality: ocr → text
Parameters: 1B total · 1B active
Size: 1gb

Stable

DeepSeek OCR v2

DeepSeek AI · 2026-01-27

Modality: ocr → text
Parameters: 3B total · 0.6B active
Size: 7gb

Experimental

Trinity Large

Arcee AI · 2026-01-27

Modality: text, code → text, tool, code
Parameters: 398B total · 13B active
Size: 376gb

Stable

Kimi K2.5

Moonshot AI · 2026-01-26

Modality: text, code, image → text, code, tool
Parameters: 1058B total · 32B active
Size: 550gb

Stable

GLM 4.7

Z AI · 2025-12-22

Modality: text, code → text, tool, code
Parameters: 358B total · 32B active
Size: 203gb

Experimental

Devstral 2

Mistral · 2025-12-08

Modality: code → code, tool
Parameters: 123B total · 123B active
Size: 119gb

Stable

Mistral Large 3

Mistral · 2025-12-01

Modality: text, image, code → text, code, tool
Parameters: 673B total · 41B active
Size: 375gb

Stable

DeepSeek V3.2

DeepSeek AI · 2025-11-30

Modality: text → text, tool
Parameters: 685B total · 37B active
Size: 642gb

Stable

FLUX.2 Dev

Black Forest Labs · 2025-11-25

Modality: text → image
Parameters: 32B total · 32B active
Size: 60gb

Stable

Kimi K2 Thinking

Moonshot AI · 2025-11-06

Modality: text, code → text, code
Parameters: 1058B total · 32B active
Size: 553gb

Stable

GLM 4.6

Z AI · 2025-09-30

Modality: text → text, tool, code
Parameters: 200B total · 32B active
Size: 187gb

Stable

Qwen 3 VL

Alibaba Cloud · 2025-09-23

Modality: image, text → text
Parameters: 235B total · 22B active
Size: 125gb

Stable

Qwen 3 Next Thinking

Alibaba Cloud · 2025-09-10

Modality: text → text
Parameters: 80B total · 3B active
Size: 44gb