Liquid AI
Company www.liquid.ai ● Covered in 12 stories + Follow ✴ AI Graph
Liquid AI is a company focused on deploying open AI models for on-device inference and tooling. Recent coverage includes releases of LFM2.5-2.6B for local agentic tool-calling with 128K context, LFM2.5-VL-3B for on-device vision-language screen understanding and tool calls (with multiple deployment formats), and LFM2.5 draft/speculative-decoding checkpoints via DSpark. The company has also published open-source benchmarking through Pipette, positioning evaluation around reproducible on-device unit configurations such as model plus quantization, runtime, and hardware.
Updated 16 September 2026
Signals
4 stories (-50%)
Media momentum
As of 17 Sep 2026 · Visible stories in the last 30 days, compared with the 30 days before.
4 sources
Source diversity
As of 17 Sep 2026 · Distinct publications behind this entity's visible coverage.
4 events
Release activity
As of 17 Sep 2026 · Model, product and open-source release events in the last 90 days whose coverage involves this entity.
—
Funding signals
As of 17 Sep 2026 · Funding and acquisition events in the last 90 days whose coverage involves this entity.
Jul 2026 → Aug 2026
Coverage span
As of 17 Sep 2026 · First to most recent month of TLDRocket coverage of this entity.
Management
updated 2 Sep 2026- Ramin Hasani Co-founder And CEO
- Rachel Link Robinson Chief People Officer
- Kelly Robinson Executive Business Partner to the C-suite
- Justin Jow CFO
- Jeffrey Li COO
- Paul Sieminski Chief Legal Officer
- Alexander Amini Co-founder & Chief Science Officer
- Daniela Rus Director
- Felix Naser COO
- Mathias Lechner Chief Technology Officer
Recent changes
- Caroline Hoogland left (was Marketing & Comms Lead)
- Jeffrey Li Head of Product/GTM → COO
Latest developments
Liquid AI Open-Sources Pipette: A Reproducible Benchmarking Suite That Measures On-Device Models, Quantization, Runtime and Hardware Together
MarkTechPost · 3 weeks ago ·
34
Liquid AI Releases LFM2.5-DSpark Draft Models That Deliver Up to 3.18x Faster Decoding Without Changing Model Outputs
MarkTechPost · 3 weeks ago ·
37
Up to 3.2x Faster Inference with LFM2.5-DSpark
Hugging Face · 3 weeks ago ·
30
LFM2.5 Q4\_0 Checkpoints from Quantization-Aware Distillation
Hugging Face · 4 weeks ago ·
36
Liquid AI Releases LFM2.5-VL-3B: A 3B Vision-Language Model That Reads Screens, Grounds Objects, and Calls Tools On-Device
MarkTechPost · 1 month ago ·
26
LFM2.5-VL-3B for Better and Faster Vision Capabilities for the Edge
Hugging Face · 1 month ago ·
22
Nativ Runs AI Models Locally on Apple Silicon
GitHub Pages · 1 month ago ·
21
Liquid AI Releases LFM2.5-2.6B: An On-Device Agentic Model With 128K Context, Tool Calling, And Open Weights
MarkTechPost · 1 month ago ·
45
August 2026
Liquid AI released DSpark draft model checkpoints for three LFM2.5 models, enabling speculative decoding to speed up inference Model release
MacPaw and Liquid AI announce a partnership Partnership
Liquid AI releases LFM2.5-2.6B, an open-weight 2.6B parameter language model optimized for on-device agentic AI tasks Open source release
- Liquid AI Open-Sources Pipette: A Reproducible Benchmarking Suite That Measures On-Device Models, Quantization, Runtime and Hardware Together
- Liquid AI Releases LFM2.5-DSpark Draft Models That Deliver Up to 3.18x Faster Decoding Without Changing Model Outputs
- Up to 3.2x Faster Inference with LFM2.5-DSpark
- LFM2.5 Q4\_0 Checkpoints from Quantization-Aware Distillation
- Liquid AI Releases LFM2.5-VL-3B: A 3B Vision-Language Model That Reads Screens, Grounds Objects, and Calls Tools On-Device
- LFM2.5-VL-3B for Better and Faster Vision Capabilities for the Edge
- Nativ Runs AI Models Locally on Apple Silicon
- Liquid AI Releases LFM2.5-2.6B: An On-Device Agentic Model With 128K Context, Tool Calling, And Open Weights
- MacPaw taps Liquid AI to offer on-device inference to devs building for its app store
- Deploy local agents everywhere with LFM2.5-2.6B
July 2026
Liquid AI releases LFM2.5-Encoder models for efficient long-context CPU inference Model release
Relationships
Products & technology
- Develops LFM2.5-Encoder-230M · 2 sources
- Develops LFM2.5-Encoder-350M · 2 sources
- Develops LFM2.5-2.6B · 2 sources
- Develops LFM2.5-VL-3B · 2 sources
- Develops Elix · 1 source
- Supplies Nativ · 1 source
- Develops Pipette · 1 source
- Develops LFM2.5-DSpark · 1 source
- Deploys LFM2.5-DSpark · 1 source
- Develops LFM2.5 · 1 source
- Supplies QAD Q4_0 GGUF checkpoints · 1 source
Partnerships
- MacPaw partnered with this company · 1 source