LFM2.5-1.2B-Instruct
LFM2.5-1.2B-Instruct is behind the leaders in instruction following. Too few results yet to rate reasoning, coding, agentic tasks, safety, long context, math, multimodal tasks, multilingual tasks, or factuality.
Capability profile
Bars show the model's median result as a share of the leading model's, per capability. Select a row to see its results.
Too few results yet to rate Reasoning, Coding, Agentic, Safety, Long Context, Math, Multimodal, Multilingual or Factuality.
Price
No current price is tracked for this model. See the rate card
Evidence
44results on34benchmarks
- 15 aggregator
- 12 vendor-reported
- 17 cross-referenced
From 5 sources · latest Oct 16, 2026 · How verification works
LFM2.5-1.2B-Instruct benchmark results
44 results on 34 benchmarks, grouped by capability. Each bar is the result as a share of the capability leader's; every result links to its source.
42.0% behind the leader1 of 3 ranked benchmarks measured
- IFBench43.81Oct 16, 2026aa_ifbench
0 of 6 ranked benchmarks measured
- 0.00Oct 16, 2026
- GPQA Diamond32.63Oct 16, 2026gpqa
- Humanity's Last Exam6.72Oct 16, 2026aa_hle
Show 2 more reasoning resultsHide 2 reasoning results
- GPQA Diamond38.89Aug 9, 2026GPQA
- 38.89Aug 9, 2026
0 of 10 ranked benchmarks measured
- Terminal-Bench Hard0.00Oct 16, 2026aa_terminalbench_hard
- SciCode2.31Sep 4, 2026aa_scicode
0 of 7 ranked benchmarks measured
- 0.00Jun 15, 2026
0 of 3 ranked benchmarks measured
- 0.00Oct 16, 2026
0 of 4 ranked benchmarks measured
- AA-Omniscience · Accuracy7.02Oct 16, 2026omniscienceAccuracy
- AA-Omniscience · Non-hallucination14.91Oct 16, 2026omniscienceNonHallucination
More results
Benchmarks outside the capability baskets. They are not ranked against a leader.
- AA Intelligence5.21Oct 16, 2026aa_intelligence_index
- -72.10Oct 16, 2026
- τ²-Bench Telecom (AA run)10.82Oct 16, 2026aa_tau2
- 0.77Jun 18, 2026
- 3.61Jun 18, 2026
- MATH500 (Pass@1)6.02Aug 20, 2026MATH500
Show 24 more resultsHide 24 results
- AIME25 no tools14.00Aug 9, 2026AIME25
- AIME25 no tools14.00Aug 9, 2026AIME25
- Avg.50.10Aug 9, 2026Avg
- 49.12Aug 9, 2026
- 49.12Aug 9, 2026
- 39.68Aug 9, 2026
- 4.34Aug 20, 2026
- 64.52Aug 9, 2026
- 5.31Aug 20, 2026
- 86.23Aug 9, 2026
- 86.23Aug 9, 2026
- 46.29Aug 9, 2026
- 50.20Aug 9, 2026
- 50.00Aug 9, 2026
- 31.72Aug 9, 2026
- 28.66Aug 9, 2026
- 47.61Aug 9, 2026
- 31.42Aug 9, 2026
- MATH-500 (EM)5.78Aug 20, 2026MATH-500
- MATH-500 (EM)63.20Aug 9, 2026MATH-500
- 5.52Aug 20, 2026
- 44.35Aug 9, 2026
- 44.35Aug 9, 2026
- 3.90Aug 20, 2026
LFM2.5-1.2B-Instruct: common questions
Who makes LFM2.5-1.2B-Instruct?
LFM2.5-1.2B-Instruct is made by Liquid AI.
When was LFM2.5-1.2B-Instruct released?
LFM2.5-1.2B-Instruct was released on Jan 5, 2026, according to Artificial Analysis.
What is LFM2.5-1.2B-Instruct good at?
LFM2.5-1.2B-Instruct is behind the leaders in instruction following. Too few results yet to rate reasoning, coding, agentic tasks, safety, long context, math, multimodal tasks, multilingual tasks, or factuality.
How many benchmarks has LFM2.5-1.2B-Instruct been tested on?
We track 44 results for LFM2.5-1.2B-Instruct on 34 benchmarks from 5 sources. The latest was recorded on Oct 16, 2026.
About this record
Where LFM2.5-1.2B-Instruct's numbers come from, and every name it appears under.
- Tracked since
- Jul 20, 2026
- Newest source mention
- Aug 20, 2026
Where the results come from
Verification: 44 scores · 0 independently verified · 15 aggregator-attributed · 17 vendor cross-reference · 12 vendor-reported. How these tiers are assigned
From 5 sources on 2 sites. Hugging Face supplies 29 of them. Bars are coloured by trust tier.
- huggingface.co29
- artificialanalysis.ai15