SE
SenseTime
Multimodal model · China ·Apr 2026

SenseNova U1

Multimodal Open Source
8B
Parameters
32.8K
Context

About SenseNova U1

SenseNova U1 is an AI model developed by SenseTime, in the multimodal category, released in 2026, made available as an open-source model with 8B params. A native unified multimodal model series on SenseTime's NEO-unify architecture, which eliminates the separate visual enc.

On this page you'll find SenseNova U1's full specifications, including a context window of 32.8K tokens. Review provider pricing and benchmark scores below, or compare SenseNova U1 head-to-head with other multimodal models.

→ More models by SenseTime → All multimodal models → AI benchmarks leaderboard

Run SenseNova U1 locally — what to buy

Share this build ↗

Sized to what SenseNova U1 actually needs (~33k context at Q4), not to the biggest GPU: Good = cheapest that runs it, Better = best value, Best = most headroom. Speed is a hardware estimate; anything past ~120 tok/s is shown as instant.

Good
Intel Arc B580 12GB
Consumer GPU · ~$249
~73.3 tok/s
Q4 · fits full context · up to 33k ctx
Runs with: ExLlamaV2, KoboldCpp, LM Studio
See full estimate →
Better
Intel Arc A770 16GB
Consumer GPU · ~$329
~90 tok/s
Q4 · fits full context · up to 33k ctx
Runs with: ExLlamaV2, KoboldCpp, LM Studio
See full estimate →
Best
NVIDIA GeForce RTX 5070
Consumer GPU · ~$549
~120 tok/s
Q4 · fits full context · up to 33k ctx
Runs with: ExLlamaV2, KoboldCpp, LM Studio
See full estimate →

Software support

·ExLlamaV2 ·KoboldCpp ·LM Studio ·MLX ·Ollama ·SGLang ·TensorRT-LLM ·llama.cpp ·vLLM ?KTransformers ?text-generation-webui

✓ measured · · compatible · — not supported. Informational only — speed is hardware-based.

Consumer GPU
36–120+ tok/s
32 options →
Apple
21.8–120+ tok/s
18 options →
Datacenter GPU
120+ tok/s
13 options →

Community

Would you run SenseNova U1 again?
No votes yet — be the first.
Did it run? — community reports by hardware

No reports yet. Own this model on some hardware? Be the first to confirm it runs.

Discussion

No comments yet. Start the discussion.

More models like SenseNova U1

AI
NeuroChess
· 72.3K
 SE
SenseNova V6
SenseTime · 600B
 SE
SenseNova Unified Large Model
SenseTime
 SE
SenseChat 5.5
SenseTime · 600B
 SE
SenseChat 5.0
SenseTime · 600B
 SE
SenseChat-DataAnalysis V4
SenseTime

Frequently asked questions

What is SenseNova U1? ▶

A native unified multimodal model series on SenseTime's NEO-unify architecture, which eliminates the separate visual enc It is tracked on GenAIList with its specifications, benchmark scores and provider pricing.

Who created SenseNova U1? ▶

SenseNova U1 was developed by SenseTime and released in 2026.

Is SenseNova U1 open source or proprietary? ▶

SenseNova U1 is released as open source — its weights are publicly available to download, run locally and fine-tune.

What is SenseNova U1's context window? ▶

SenseNova U1 supports a context window of 32.8K tokens, and has 8B params.

How much does SenseNova U1 cost? ▶

Pricing for SenseNova U1 depends on the provider. See the providers table on this page for the latest API rates.

How does SenseNova U1 perform on benchmarks? ▶

Benchmark scores for SenseNova U1 are listed on this page as they are published. You can compare it head-to-head with other models on the GenAIList compare tool.

SenseNova U1: The Open-Source Model That Thinks in Images
SenseNova U1 First Test – A New Kind of Open Source Image Model!
SenseNova-U1: Unified Vision and Generation
SenseNova-U1 | LOW VRAM Workflow

Know a great video about SenseNova U1?