LFM2.5-2.6B
LFM2.5-2.6B is behind the leaders in instruction following and factuality. Too few results yet to rate reasoning, coding, agentic tasks, safety, long context, math, multimodal tasks, or multilingual tasks.
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 or Multilingual.
Price
No current price is tracked for this model. See the rate card
Evidence
30results on30benchmarks
- 14 aggregator
- 16 vendor-reported
From 4 sources · latest Oct 7, 2026 · How verification works
API features
Tool callingStructured outputsReasoning
As listed by OpenRouter
LFM2.5-2.6B benchmark results
30 results on 30 benchmarks, grouped by capability. Each bar is the result as a share of the capability leader's; every result links to its source.
34.2% behind the leader1 of 3 ranked benchmarks measured
- 59.17Oct 6, 2026
37.1% behind the leader2 of 4 ranked benchmarks measured
- AA-Omniscience · Non-hallucination84.00Oct 7, 2026omniscienceNonHallucination
- AA-Omniscience · Accuracy4.35Oct 7, 2026omniscienceAccuracy
0 of 6 ranked benchmarks measured
- 0.00Oct 7, 2026
- GPQA Diamond55.76Oct 7, 2026gpqa
- Humanity's Last Exam6.21Oct 7, 2026aa_hle
0 of 10 ranked benchmarks measured
- SciCode14.35Oct 7, 2026aa_scicode
- Terminal-Bench 2.14.49Oct 7, 2026terminalbenchV21
- 59.41Oct 6, 2026
0 of 7 ranked benchmarks measured
- 0.00Oct 7, 2026
- τ-Bench V3 · Banking7.22Oct 7, 2026tauBanking
0 of 3 ranked benchmarks measured
- 5.67Oct 7, 2026
More results
Benchmarks outside the capability baskets. They are not ranked against a leader.
- AA Intelligence8.39Oct 7, 2026aa_intelligence_index
- -10.95Oct 7, 2026
- Artificial Analysis Coding Index7.74Sep 9, 2026aa_coding_index
- 2.41Sep 4, 2026
- Claw Eval (pass@3)62.85Oct 6, 2026Claw-Eval
- τ³-Bench Banking5.67Oct 6, 2026Tau3 Banking
Show 12 more resultsHide 12 results
- 51.87Oct 6, 2026
- AIME25 no tools51.87Aug 10, 2026AIME25
- 56.88Oct 6, 2026
- 26.89Aug 10, 2026
- Claw-Eval Avg62.85Aug 10, 2026Claw-Eval average (EN)
- 4.32Aug 30, 2026
- 4.54Aug 30, 2026
- LiveCodeBenchV6 no tools59.41Aug 10, 2026LiveCodeBenchv6
- MATH500 (Pass@1)5.42Aug 30, 2026MATH500
- 4.71Aug 30, 2026
- 5.07Aug 30, 2026
- 68.22Oct 6, 2026
LFM2.5-2.6B: common questions
Who makes LFM2.5-2.6B?
LFM2.5-2.6B is made by Liquid AI.
When was LFM2.5-2.6B released?
LFM2.5-2.6B was released on Aug 4, 2026, according to Liquid AI's own announcement.
What is LFM2.5-2.6B good at?
LFM2.5-2.6B is behind the leaders in instruction following and factuality. Too few results yet to rate reasoning, coding, agentic tasks, safety, long context, math, multimodal tasks, or multilingual tasks.
How many benchmarks has LFM2.5-2.6B been tested on?
We track 30 results for LFM2.5-2.6B on 30 benchmarks from 4 sources. The latest was recorded on Oct 7, 2026.
Which API features does LFM2.5-2.6B support?
OpenRouter lists tool calling, structured outputs, and reasoning for LFM2.5-2.6B.
About this record
Where LFM2.5-2.6B's numbers come from, and every name it appears under.
- Tracked since
- Aug 4, 2026
- Newest source mention
- Aug 20, 2026
Where the results come from
Verification: 30 scores · 0 independently verified · 14 aggregator-attributed · 16 vendor-reported. How these tiers are assigned
From 4 sources on 3 sites. Artificial Analysis supplies 14 of them. Bars are coloured by trust tier.
- artificialanalysis.ai14
- huggingface.co9
- api.llm-stats.com7