[whatmodel.sh](../../README.md) / LFM3 1.7B # LFM3 1.7B Liquid AI · released 2026-10 · 1.73B params · lfm1.0 **score 35.0** ±15.0 · #2 of 2 · from 7 benchmark results across 1 source · † no independently run results verified Score within the [GPQA Diamond](https://whatmodel.sh/?set=gpqa-diamond%40first-party-liquid-scores&score=set) comparison set only. ## What it's good at ● marks LFM3 1.7B among every model on that task or test, worst to best. 
task            vs the others            rank  counts  tests
agent tasks     ●··········█         #2 of 2*     31%      6
coding          ●··········█         #2 of 2*     10%      4
chat & writing  ●··········█         #2 of 2*    8.9%      1
web search                     not ranked yet       —      1
agent tasks · #2 of 2 · 31% of its score 
test                                result  vs the others in its run   difficulty        counts  why
WebArena-Verified (maker)            38.4%  █··●···█·█·█ beats 1/4     hard · avg 32%       27%  sharp · small run
SWE-bench Verified (maker)           43.4%  ▁▁▁▁▁▁▂●█▆▁▁ beats 33/138  medium · avg 47%    4.1%  at its level · small run
SWE-bench Multilingual (maker)       23.0%  ▁··●▃▁▅██▂▂▁ beats 1/65                           —  not counted yet: no other model in the …
Terminal-Bench 2.1 (maker)           12.9%  ▁·▁·▁●▄▃▆█▅█ beats 3/63                           —  not counted yet: no other model in the …
Humanity's Last Exam with tools (…   13.9%  ▁·●▂▂▁▄▄▇█▄▅ beats 1/62                           —  not counted yet: no other model in the …
MCP-Atlas (public set) (maker)       32.6%  ▂·▁▂●·▁▃█▆▆▂ beats 5/39                           —  not counted yet: no other model in the …
coding · #2 of 2 · 10% of its score 
test                                result  vs the others in its run   difficulty        counts  why
SciCode (level unspecified) (make…   28.5%  ▁····▃▃●▅█▆▂ beats 11/40   hard · avg 30%      5.8%  at its level · small run
SWE-bench Verified (maker)           43.4%  ▁▁▁▁▁▁▂●█▆▁▁ beats 33/138  medium · avg 47%    4.1%  at its level · small run
LiveCodeBench v6 (window unspecif…   73.5%  ▂▃▄▄▅██●▄▇▄▆ beats 40/67                          —  not counted yet: no other model in the …
SWE-bench Multilingual (maker)       23.0%  ▁··●▃▁▅██▂▂▁ beats 1/65                           —  not counted yet: no other model in the …
chat & writing · #2 of 2 · 8.9% of its score 
test                                result  vs the others in its run    difficulty           counts  why
IFEval (aggregation unspecified) …   96.7%  ▁▂▄▂▂▄▆██▅▂● beats 144/145  very easy · avg 97%    8.9%  maxed out · small run
web search · 0.0% of its score 
test                                result  vs the others in its run  counts  why
Humanity's Last Exam with tools (…   13.9%  ▁·●▂▂▁▄▄▇█▄▅ beats 1/62        —  not counted yet: no other model in the …
other tests (5) 
test                                result  vs the others in its run    difficulty            counts  why
Humanity's Last Exam (maker)          7.2%  ▁▁▁●▆▆▇██▆▁▃ beats 8/111    very hard · avg 7.0%    8.9%  too hard for it · small run
IFBench (aggregation unspecified)…   73.0%  ▂▆▂▄▄▄▂▄▄●█▇ beats 62/86    easy · avg 74%          8.0%  at its level · small run
GPQA Diamond (maker)                 63.4%  ▁▁▁▁▂▄▅●███▄ beats 124/322  medium · avg 69%        0.8%  blunt · at its level · small run
RULER 128k (maker)                   65.6%  ▃··▅▄●▅█▇·▃▃ beats 12/33                               —  not counted yet: no other model in the …
AutomationBench (Public) (maker)     41.3%  █····▄····▄● beats 4/4                                 —  not counted yet: no other model in the …
## How the score was worked out Each result implies a score on its own, after adjusting for how hard that test is and who ran it. The model's score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its *counts* column above; here it is added up by who ran the tests. [Method](../method.md) 
source                                    results  share of the score
Liquid AI own-model benchmark facts (su…        7   64% ▇▇▇▇▇▇▇▇
prior: pull toward the middle                       36% ▇▇▇▇
weighted average                                   = 35.0
Not counted (7) 
source                  test                            score  why
Liquid AI own-model b…  AutomationBench (Public)        41.3%  too few models have it yet
Liquid AI own-model b…  MCP-Atlas (public set)          32.6%  too few models have it yet
Liquid AI own-model b…  SWE-bench Multilingual          23.0%  too few models have it yet
Liquid AI own-model b…  Terminal-Bench 2.1              12.9%  too few models have it yet
Liquid AI own-model b…  LiveCodeBench v6 (window unsp…  73.5%  too few models have it yet
Liquid AI own-model b…  Humanity's Last Exam with too…  13.9%  too few models have it yet
Liquid AI own-model b…  RULER 128k                      65.6%  too few models have it yet
## How to run it ### llama.cpp (recommended) 1. Install llama.cpp. Homebrew and winget installs name the commands llama-cli and llama-server instead. [Other ways to install](https://github.com/ggml-org/llama.cpp/blob/master/docs/install.md) ```sh # macOS / Linux curl -LsSf https://llama.app/install.sh | sh # Windows irm https://llama.app/install.ps1 | iex ``` 2. Run LFM3 1.7B (Q4_K_M, needs ~1.04 GB, from [huggingface.co/LiquidAI](https://huggingface.co/LiquidAI/LFM3-1.7B-GGUF/blob/bad5729bba61d076fff20d2548e62526665af215/README.md)) ```sh llama cli -hf LiquidAI/LFM3-1.7B-GGUF:Q4_K_M ``` 3. Or start a local server and open the address it prints (http://localhost:9931) ```sh llama serve -hf LiquidAI/LFM3-1.7B-GGUF:Q4_K_M ``` ### LM Studio 1. Install LM Studio. Open the app once before using the lms command. [Download it (macOS / Windows / Linux)](https://lmstudio.ai/download) 2. Run LFM3 1.7B (Q4_K_M, needs ~1.04 GB, from [huggingface.co/LiquidAI](https://huggingface.co/LiquidAI/LFM3-1.7B-GGUF/blob/bad5729bba61d076fff20d2548e62526665af215/README.md)) ```sh lms get https://huggingface.co/LiquidAI/LFM3-1.7B-GGUF@Q4_K_M -y ``` 3. Then run lms ls to see its name, and lms chat followed by that name ### Ollama 1. Install Ollama. [Download it (macOS / Windows)](https://ollama.com/download) ```sh # Linux curl -fsSL https://ollama.com/install.sh | sh ``` 2. Run LFM3 1.7B (Q4_K_M, needs ~1.04 GB, from [huggingface.co/LiquidAI](https://huggingface.co/LiquidAI/LFM3-1.7B-GGUF/blob/bad5729bba61d076fff20d2548e62526665af215/README.md)) ```sh ollama run hf.co/LiquidAI/LFM3-1.7B-GGUF:Q4_K_M ``` Model repository and documentation: [huggingface.co/LiquidAI](https://huggingface.co/LiquidAI/LFM3-1.7B) [interactive version](https://whatmodel.sh/model/lfm3-1-7b) · [JSON](https://whatmodel.sh/api?format=json&all=1)