LFM2.5-8B-A1B

LFM2.5-8B-A1B is a reasoning model from LiquidAI. 15 benchmarks count toward its score, in 6 categories.

availableShows if the model has enough results for an index.
IndexOverall score out of 100.32.0 ±7.2
CoverageShare of the index weight with results.70%
SpeedOutput tokens per second.335/s
Input / 1MUS dollars per 1M input tokens.Free
Output / 1MUS dollars per 1M output tokens.Free
ContextMaximum tokens in one request.128K
EloLMArena rating and rank.N/A

The index is a score out of 100. The ± range shows how much it can change.

CapabilitiesScore per category, out of 100.

Out of 100
AgenticMulti-step tasks with tools.
48.1
CodingCode writing and repair.
N/A
ReasoningLogic problems and puzzles.
23.3
MultimodalTasks with images and text.
N/A
KnowledgeFacts and expert knowledge.
27.5
MultilingualTasks in many languages.
28.5
InstructionTasks with strict rules in the prompt.
42.2
MathMath problems.
24.5

Results

15 counted
BenchmarkThe test name.CategoryThe capability that the test measures.ResultThe score from the publisher.IndexThis result as a score out of 100.RunThe settings of the run.DateDate of the result.Published byThe source of the result.
Instruction-Following EvalInstruction91.8%49.2Jeffrey Zhou et al.
MATH-500 Problem SetMath88.8%41.0Dan Hendrycks et al.
τ²-Bench Tool-Agent-User EvaluationAgentic88.1%62.0Victor Barres et al.
Instruction Following BenchmarkInstruction56.5%31.8Benchmark authors
Artificial Analysis IFBenchInstruction55.6%45.5Artificial Analysis
Artificial Analysis GPQA DiamondKnowledge51.3%21.1Artificial Analysis
AIME 2026Math50.0%20.6Qwen
Berkeley Function Calling Leaderboard v4Agentic49.7%34.1Arcee AI
American Invitational Mathematics Examination 2025Math42.5%12.0Mathematical Association of America
EuroEval FrenchMultilingual29.6%39.3EuroEval
EuroEval SpanishMultilingual24.1%32.5EuroEval
EuroEval PortugueseMultilingual22.0%29.8EuroEval
EuroEval DutchMultilingual21.1%28.6EuroEval
EuroEval ItalianMultilingual20.8%28.3EuroEval
EuroEval GermanMultilingual20.6%28.0EuroEval
EuroEval SwedishMultilingual14.3%20.2EuroEval
EuroEval PolishMultilingual10.8%15.8EuroEval
Artificial Analysis Omniscience AccuracyKnowledge9.4%25.5Artificial Analysis
Artificial Analysis Intelligence IndexKnowledge7.2%31.4Artificial Analysis
Artificial Analysis Humanity's Last ExamKnowledge6.9%32.0Artificial Analysis
Artificial Analysis Long Context ReasoningReasoning0.0%8.3Artificial Analysis
Critical Physics TasksReasoning0.0%38.4Artificial Analysis

15 benchmarks count, from 22 of 22 results. A grey row does not count. Too few models took that benchmark.

Sources

BenchLM benchmark aggregationUsed with attribution; per-benchmark results credited to their original authorsEuroEval, collected directlyMIT — the leaderboard site and its CSV routes are in the licensed repository

More from LiquidAI

LFM2.5-2.6B27.5LFM2.5-VL-1.6B-Extract19.4LFM2 1.2BUnrankedLFM2 350MUnrankedLFM2 700MUnranked