LiquidAI: LFM2.5-2.6B (free)
liquid/lfm-2.5-2.6b:free · Liquid AINot yet established- Declared family
- Unknown
- Entered catalogue
- 11 August 2026
- Last tested
- 21 August 2026
- Measured signature
- tk_2c236574
Identity statement
LiquidAI: LFM2.5-2.6B (free) has a clean measured fingerprint that currently sits alone: no other measured model shares its signature.
What we checked
Four different kinds of evidence, and what each one showed| We checked | What we found | Where it came from |
|---|---|---|
| How it counts tokens | Counted cleanly, and no other model in the catalogue counts the same way | tk_2c236574 |
| How its API is set up | A combination of settings no other listing in the catalogue uses | api_d2ad76e1 |
| How much it can read and write | Reads up to 65,536 tokens at once, writes up to 8,192 | its own listing |
| Whether it thinks before answering | Always reasons, and cannot be turned off | its own listing |
| Who served our requests | Liquid | we measured it, 21 August 2026 |
Closest measured models
Agreement across the mutually clean test stringsDeepSeek: DeepSeek V4 Pro 0813deepseek/deepseek-v4-pro-081329/50 · 58%DeepSeek: DeepSeek V4 Flash Vision Expdeepseek/deepseek-v4-flash-vision-exp29/50 · 58%DeepSeek: DeepSeek V4 Flash 0731deepseek/deepseek-v4-flash-073129/50 · 58%DeepSeek: DeepSeek V4 Flash 0423deepseek/deepseek-v4-flash29/50 · 58%StepFun: Step 3.5 Flashstepfun/step-3.5-flash29/50 · 58%StepFun: Step 3.7 Flashstepfun/step-3.7-flash29/50 · 58%
Click a row to open the full comparison.
The measured fingerprint
What each test string cost this model, in tokensWe sent LiquidAI: LFM2.5-2.6B (free) fifty short pieces of text and recorded what each one cost it in tokens. The bars below are those costs. Two models built on the same tokenizer produce the same bars; a model built on a different one produces a different set, which is what makes this a fingerprint.
All 50 rows
English and whitespace
en-prose14
en-long18
spaces-203
spaces-603
tabs-204
newlines-204
mixed-ws7
Digits
digits-93
digits-124
digits-3010
digits-sep7
float-long9
CJK
zh-common12
zh-long16
zh-rare18
ja-kana9
ja-kanji11
ko11
Other scripts
ru16
ar10
he14
hi14
th14
el17
Emoji
emoji-basic10
emoji-skin20
emoji-zwj-family11
emoji-zwj-x333
emoji-flags16
emoji-prof19
Rare Unicode
math27
boxdraw29
combining12
cjk-ext-b16
surrogates40
zalgo36
rtl-mix7
Code
py-code25
py-indent15
json20
html16
regex49
camel6
snake7
Repetition and encodings
rare-word-x531
repeat-tok41
base6431
hex13
url18
uuid27
Overhead subtracted: 10 prompt tokens.
Declared record
What the catalogue claims about this modelLFM2.5-2.6B is a compact reasoning model from Liquid AI. It is suited for agent workflows, data extraction, RAG, and long-context processing. Liquid advises against using it for agentic coding or...
- Modalities
- text->text
- Declared tokenizer
- Other
- Prompt price
- $0 / M tokens
- Completion price
- $0 / M tokens
Supported parameters
frequency_penaltyinclude_reasoninglogprobsmax_completion_tokensmax_tokensmin_ppresence_penaltyreasoningrepetition_penaltyresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_ktop_logprobstop_p
defaults: {"temperature":0.1,"top_k":50,"repetition_penalty":1.1}
History
Every observation, kept as taken- 11 August 2026Enters the OpenRouter catalogue with no declared family.
- 21 August 2026Fingerprinted in the YFarmX catalogue sweep · 50 of 50 strings measured clean.
CompareCompare now