Skip to content

Liquid AI · Chat

LFM2-8B-A1B

Liquid AI 🇺🇸ChatLFM2 family
llminfo IQ
≈ 106
Market median price
$0.1125 blended / 1M
Max context
33K
Listings
2

Evidence and sources

🤗 LiquidAI/LFM2-8B-A1Bllminfo IQ ≈ 106, ranked among chat models. From 8 benchmarks. Little data behind it, so treat it as a rough estimate.How llminfo IQ is computedGood value (52.9). Cheaper per task than its benchmark results predict. From 1 benchmark with measured task costs. Little data behind it, so treat it as a rough estimate.How Value for money worksReleased 7 Oct 2025Maker HQ: US (source)
Sources: Core42 Compass Model Pricing, Hugging FacePrimary data from Core42 Compass Model Pricing from Hugging Face: release date

Advantages and trade-offs

Compared with 10 chat models, picked mainly by popularity and IQ

Trade-offs

We only make a claim when there's enough data on models of the same type. Updated 22 Nov, 04:34 UTC.Prices come from the model's maker where they sell it, otherwise their official partner, otherwise the market median. This comparison uses 6 maker's prices; 4 market median prices.

Input & output

from the model's type
Text
Text

Listings

 Core42LFM2-8B-A1BSource: Core42 Compass Model PricingNot published by the provider$0.10$0.15

1 listing without a price is hidden.

Benchmarks

Named effort levels run from lowest to highest. Select a result to see its source details.

Knowledge

Math

Reasoning

Other

Data credits: Hugging Face official benchmark leaderboards; licence varies by benchmark · SnakeBench (Greg Kamradt / ARC Prize Foundation), MIT (licence)

Around the web

  • Introducing LFM2: The Fastest On-Device Foundation Models on the Market — Blog — Liquid AI
    outperforms Gemma 3 1B IT, and our tiniest checkpoint, LFM2-350M, is competitive with Qwen3-0.6B and …outperforms Gemma 3 1B IT, and our tiniest checkpoint, LFM2-350M, is competitive with Qwen3-0.6B and Llama 3.2 1B Instruct. Quick picks on LFM2 Fast training & inference LFM2 achieves 3x faster train…

    liquid.ai

  • LFM2-Audio: An End-to-End Audio Foundation Model — Blog — Liquid AI
    generation in one compact system, it enables conversational AI on devices where speed, privacy, and …generation in one compact system, it enables conversational AI on devices where speed, privacy, and efficiency matter most. With LFM2-Audio, we’re taking another step toward multimodal agents that ru…

    liquid.ai

  • LFM2-2.6B - Liquid Docs
    LFM2 (Dense) Runs on phones and laptops Native function calling support TRL compatible (SFT, DPO, …LFM2 (Dense) Runs on phones and laptops Native function calling support TRL compatible (SFT, DPO, GRPO) Quick Start model.generate(**inputs, ${samplingParams}max_new_tokens=512) input_length = inputs…

    docs.liquid.ai

  • LFM2-350M

    model.generate(**inputs, ${samplingParams}max_new_tokens=512) input_length = inputs["input_ids"].shape[1] response = tokenizer.decode(output[0][input_length:], skip_special_tokens=True) print(respons…

    docs.liquid.ai

  • LFM2: Advancing Open Science in AI — Blog — Liquid AI
    Models Research Engineering Blog Developers Solutions Company Connect News LFM2: Advancing Open Science in AI …Models Research Engineering Blog Developers Solutions Company Connect News LFM2: Advancing Open Science in AI News DEC 1, 2025 LFM2: Advancing Open Science in AI At Liquid AI, we believe that scienti…

    liquid.ai

  • From Cloud to AI PC: Launching Liquid’s Largest LFM2 Model Alongside Our Growing Partner Ecosystem — Blog — Liquid AI
    Models Research Engineering Blog Developers Solutions Company Connect News From Cloud to AI PC: Launching …Models Research Engineering Blog Developers Solutions Company Connect News From Cloud to AI PC: Launching Liquid’s Largest LFM2 Model Alongside Our Growing Partner Ecosystem News FEB 24, 2026 From Cl…

    liquid.ai

  • LFM2-1.2B - Liquid Docs

    model.generate(**inputs, ${samplingParams}max_new_tokens=512) input_length = inputs["input_ids"].shape[1] response = tokenizer.decode(output[0][input_length:], skip_special_tokens=True) print(respons…

    docs.liquid.ai

  • LFM2-8B-A1B - Liquid Docs

    model.generate(**inputs, ${samplingParams}max_new_tokens=512) input_length = inputs["input_ids"].shape[1] response = tokenizer.decode(output[0][input_length:], skip_special_tokens=True) print(respons…

    docs.liquid.ai

  • LFM2-24B-A2B: Scaling Up the LFM2 Architecture — Blog — Liquid AI
    the LFM2 architecture scales effectively to larger sizes. With this release, the LFM2 family spans …the LFM2 architecture scales effectively to larger sizes. With this release, the LFM2 family spans nearly two orders of magnitude: from LFM2-350M to LFM2-24B-A2B. Each step up in scale has brought co…

    liquid.ai

  • LiquidAI/LFM2-700M · Hugging Face
    LFM2 is a new generation of hybrid models developed by Liquid AI , specifically designed …LFM2 is a new generation of hybrid models developed by Liquid AI , specifically designed for edge AI and on-device deployment. It sets a new standard in terms of quality, speed, and memory efficiency…

    huggingface.co

Also known as: liquid/lfm2-8b-a1b

Like this one. Tap or hover any dotted value, label or icon to see where it comes from and how it is measured. open details when tapped or hovered. How this data works