Skip to main content

Model comparison

Kimi K2.5 vs LFM2.5-8B-A1B

Data verified

Head-to-head evidence from 15 shared benchmark results across 6 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.

Moonshot AI
59.58/100
Margin
18.3pts
← winning
41.25/100
2 category wins0 category wins

Verified leaderboard positions: Kimi K2.5 #22; LFM2.5-8B-A1B unranked

BenchAlign evidence: Kimi K2.5 supported; LFM2.5-8B-A1B estimated. Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.

Evidence parity. Kimi K2.5 and LFM2.5-8B-A1B share 15 comparable benchmark results. 2 of 8 categories are comparable. 49 results are unique to Kimi K2.5; 3 to LFM2.5-8B-A1B.

Updated July 16, 2026
Shared results
15
Kimi K2.5 only
49
LFM2.5-8B-A1B only
3
Comparable categories
2 / 8

Pick Kimi K2.5 if you want the stronger benchmark profile. LFM2.5-8B-A1B only becomes the better choice if you want the cheaper token bill or you want the stronger reasoning-first profile.

Confidence note. This is a partial-evidence comparison with 15 shared benchmark results across 6 evidence categories; 2 of 8 categories currently have scoreable aggregates for both models. Treat the verdict as directional until coverage is more balanced.

Why this result

Kimi K2.5 is clearly ahead on the provisional aggregate, 61 to 37. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.

Kimi K2.5's sharpest advantage is in instruction following, where it averages 93.9 against 68.8. The single biggest benchmark swing on the page is AIME26, 95.8% to 50.0%.

Kimi K2.5 is also the more expensive model on tokens at $0.60 input / $3.00 output per 1M tokens, versus $0.00 input / $0.00 output per 1M tokens for LFM2.5-8B-A1B. That is roughly Infinityx on output cost alone. LFM2.5-8B-A1B is the reasoning model in the pair, while Kimi K2.5 is not. That usually helps on harder chain-of-thought-heavy tests, but it can also mean more latency and more token spend in real use. Kimi K2.5 gives you the larger context window at 256K, compared with 128K for LFM2.5-8B-A1B.

Category breakdown

Exact category averages are shown below. Not measured means BenchLM does not have enough sourced public coverage for that model and category.

Category scores and score margins for Kimi K2.5 and LFM2.5-8B-A1B
CategoryKimi K2.5ΔLFM2.5-8B-A1B
Inst. FollowingKimi K2.593.9Margin 25.1LFM2.5-8B-A1B68.8
MathKimi K2.560.6Margin 10.6LFM2.5-8B-A1B50.0
AgenticKimi K2.555.0MarginNo overlapLFM2.5-8B-A1BNot measured
CodingKimi K2.559.4MarginNo overlapLFM2.5-8B-A1BNot measured
ReasoningKimi K2.561.0MarginNo overlapLFM2.5-8B-A1BNot measured
KnowledgeKimi K2.557.2MarginNo overlapLFM2.5-8B-A1BNot measured
MultilingualKimi K2.582.3MarginNo overlapLFM2.5-8B-A1BNot measured
MultimodalKimi K2.578.5MarginNo overlapLFM2.5-8B-A1BNot measured

Decisive benchmark drivers

The largest measured benchmark gaps in this matchup, with exact reported values.

More
A · Kimi K2.5B · LFM2.5-8B-A1B
  1. AIME26

    Math
    Source ↗
    A 95.8%B 50.0%
    Winner: Kimi K2.5Δ 45.8
    AIME26: Kimi K2.5 scored 95.8%; LFM2.5-8B-A1B scored 50.0%. Kimi K2.5 wins this benchmark.
  2. IFEval

    Inst. Following
    Source ↗
    A 93.9%B 91.8%
    Winner: Kimi K2.5Δ 2.1
    IFEval: Kimi K2.5 scored 93.9%; LFM2.5-8B-A1B scored 91.8%. Kimi K2.5 wins this benchmark.

Operational comparison

Runtime and commercial metrics are compared only when both models have a complete sourced value.

MetricKimi K2.5LFM2.5-8B-A1BComparison
Input / output priceUSD per 1M tokensKimi K2.5$0.6 input / $3 outputLFM2.5-8B-A1B$0 input / $0 outputLFM2.5-8B-A1B has the lower combined listed price.
Generation speedtokens per secondKimi K2.545 tok/sLFM2.5-8B-A1BNot availableA complete speed comparison is not available.
First-answer latencyseconds to first tokenKimi K2.52.38 sLFM2.5-8B-A1BNot availableA complete latency comparison is not available.
Context windowmaximum listed tokensKimi K2.5256KLFM2.5-8B-A1B128KKimi K2.5 lists the larger context window.

Benchmark Deep Dive

Agentic
BenchmarkKimi K2.5LFM2.5-8B-A1BResult
Terminal-Bench 2.0Source 50.8%Not comparable
BrowseCompSource 60.6%Not comparable
Claw-EvalSource 52.3%Not comparable
QwenClawBenchSource 54.3%Not comparable
τ³-bench resultsSource 65.7%Not comparable
DeepSearchQASource 77.1%Not comparable
DeepPlanningSource 14.4%Not comparable
ToolathlonSource 27.8%Not comparable
MCP AtlasSource 29.5%Not comparable
MCP-TasksSource 59.1%Not comparable
WideResearchSource 72.7%Not comparable
τ²-bench resultsSource 95.9%16.1%Kimi K2.5 leads
APEX-Agents-AASource 11.5%Not comparable
Gert LabsSource 45.88%Not comparable
ResearchClawBenchSource 14.0%Not comparable
JobBenchSource 8.7%Not comparable
AA Agentic IndexSource 21.7%Not comparable
GDPval-AASource 25.4%Not comparable
GDPval-AASource 1009Not comparable
BFCL v4Source 49.7%Not comparable
Coding
BenchmarkKimi K2.5LFM2.5-8B-A1BResult
SWE-bench VerifiedSource 76.8%Not comparable
SWE-bench Verified*Source 70.8%Not comparable
LiveCodeBench v6Source 85.0%Not comparable
SWE-bench ProSource 50.7%Not comparable
SWE MultilingualSource 73%Not comparable
SWE-RebenchSource 58.5%Not comparable
React Native EvalsSource 77.2%Not comparable
SciCodeSource 48.7%Not comparable
Terminal-Bench HardSource 34.8%4.5%Kimi K2.5 leads
AA-SciCodeSource 49.0%7.8%Kimi K2.5 leads
AA Coding IndexSource 46.8%Not comparable
Reasoning
BenchmarkKimi K2.5LFM2.5-8B-A1BResult
LongBench v2Source 61%Not comparable
AA-LCRSource 65.3%0.0%Kimi K2.5 leads
CritPtSource 3.1%0.0%Kimi K2.5 leads
Knowledge
BenchmarkKimi K2.5LFM2.5-8B-A1BResult
GPQASource 87.6%Not comparable
GPQA-DSource 87.6%Not comparable
SuperGPQASource 69.2%Not comparable
MMLU-ProSource 87.1%Not comparable
MMLU-Pro (Arcee)Source 87.1%Not comparable
HLESource 30.1%Not comparable
Artificial Analysis Intelligence IndexSource 35.4%8.3%Kimi K2.5 leads
AA-GPQA DiamondSource 87.9%51.3%Kimi K2.5 leads
AA-HLESource 29.4%6.9%Kimi K2.5 leads
AA-Omniscience IndexSource -8.1%-33.3%Kimi K2.5 leads
AA-Omniscience AccuracySource 34.3%9.4%Kimi K2.5 leads
AA-Omniscience Hallucination RateSource 64.6%47.0%LFM2.5-8B-A1B leads
MathKimi K2.5 wins
BenchmarkKimi K2.5LFM2.5-8B-A1BResult
AIME 2025Source 96.1%42.5%Kimi K2.5 leads
AIME26Source 95.8%50.0%Kimi K2.5 leads
AIME25 (Arcee)Source 96.3%Not comparable
HMMT Feb 2025Source 95.4%Not comparable
HMMT Nov 2025Source 91.1%Not comparable
HMMT Feb 2026Source 87.1%Not comparable
MMAnswerBenchSource 81.8%Not comparable
FrontierMath v2 (Tiers 1-3)Source 27.900%Not comparable
FrontierMath v2 (Tier 4)Source 4.200%Not comparable
MATH-500Source 88.8%Not comparable
Multilingual
BenchmarkKimi K2.5LFM2.5-8B-A1BResult
MMLU-ProXSource 82.3%Not comparable
NOVA-63Source 56.0%Not comparable
Multimodal
BenchmarkKimi K2.5LFM2.5-8B-A1BResult
MMMU-ProSource 78.5%Not comparable
Video-MMESource 87.4%Not comparable
MMVUSource 80.4%Not comparable
VideoMMMUSource 86.6%Not comparable
AA-MMMU-ProSource 75.4%Not comparable
Design Arena WebsiteSource 1284Not comparable
Inst. FollowingKimi K2.5 wins
BenchmarkKimi K2.5LFM2.5-8B-A1BResult
IFEvalSource 93.9%91.8%Kimi K2.5 leads
AA-IFBenchSource 70.2%55.6%Kimi K2.5 leads
IFBenchSource 56.5%Not comparable
Frequently Asked Questions (3)

Which is better, Kimi K2.5 or LFM2.5-8B-A1B?

Kimi K2.5 is ahead on BenchLM's provisional leaderboard, 61 to 37. The biggest single separator in this matchup is AIME26, where the scores are 95.8% and 50.0%.

Which is better for math, Kimi K2.5 or LFM2.5-8B-A1B?

Kimi K2.5 has the edge for math in this comparison, averaging 60.6 versus 50. Inside this category, AIME 2025 is the benchmark that creates the most daylight between them.

Which is better for instruction following, Kimi K2.5 or LFM2.5-8B-A1B?

Kimi K2.5 has the edge for instruction following in this comparison, averaging 93.9 versus 68.8. Inside this category, AA-IFBench is the benchmark that creates the most daylight between them.

Self-host vs API cost

Estimates at 50,000 req/day · 1000 tokens/req average.

Kimi K2.5
API / mo$2,700
Self-host / mo$5,221
Break-even132M/day
LFM2.5-8B-A1B
API / mo$0
Self-host / moNot listed
Break-even
Proprietary model — self-hosting not applicable.
Model the full break-even

Related Comparisons

Last updated: July 16, 2026

The AI models change fast. We track them for you.

A weekly brief for engineers and researchers covering new models, ranking shifts, and pricing changes.

Free. No spam. Unsubscribe anytime.