Model comparison
Kimi K3 vs LFM2.5-230M
Head-to-head evidence from 2 shared benchmark results across 1 category. Overall scores shown here use the public BenchAlign v5 ranking lane.
Public leaderboard positions: Kimi K3 #4 (Supported); LFM2.5-230M unranked (Not scored). Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.
Evidence parity. Kimi K3 and LFM2.5-230M share 2 comparable benchmark results. 1 of 8 categories are comparable. 55 results are unique to Kimi K3; 4 to LFM2.5-230M.
Updated July 23, 2026- Shared results
- 2
- Kimi K3 only
- 55
- LFM2.5-230M only
- 4
- Comparable categories
- 1 / 8
Treat this as a split decision. Kimi K3 makes more sense if knowledge is the priority or you need the larger 1.05M context window; LFM2.5-230M is the better fit if you want the cheaper token bill or you would rather avoid the extra latency and token burn of a reasoning model.
Confidence note. This is a partial-evidence comparison with 2 shared benchmark results across 1 evidence category; 1 of 8 categories currently have scoreable aggregates for both models. Treat the verdict as directional until coverage is more balanced.
Why this result
Kimi K3 and LFM2.5-230M finish on the same BenchAlign overall score, so this is less about a single winner and more about where the edge shows up. The BenchAlign headline says tie; the benchmark table is where the real choice happens.
Kimi K3 is also the more expensive model on tokens at $3.00 input / $15.00 output per 1M tokens, versus $0.00 input / $0.00 output per 1M tokens for LFM2.5-230M. That is roughly Infinityx on output cost alone. Kimi K3 is the reasoning model in the pair, while LFM2.5-230M 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 K3 gives you the larger context window at 1.05M, compared with 32K for LFM2.5-230M.
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 | Kimi K3 | Δ | LFM2.5-230M |
|---|---|---|---|
| Knowledge | Kimi K361.0 | Margin← 39.8 | LFM2.5-230M21.2 |
| Agentic | Kimi K389.5 | MarginNo overlap | LFM2.5-230MNot measured |
| Multimodal | Kimi K378.5 | MarginNo overlap | LFM2.5-230MNot measured |
| Inst. Following | Kimi K3Not measured | MarginNo overlap | LFM2.5-230M50.1 |
Decisive benchmark drivers
The largest measured benchmark gaps in this matchup, with exact reported values.
More
- Source ↗
GPQA
KnowledgeA 93.5%B 25.4%Winner: Kimi K3Δ 68.1GPQA: Kimi K3 scored 93.5%; LFM2.5-230M scored 25.4%. Kimi K3 wins this benchmark.
Operational comparison
Runtime and commercial metrics are compared only when both models have a complete sourced value.
| Metric | Kimi K3 | LFM2.5-230M | Comparison |
|---|---|---|---|
| Input / output priceUSD per 1M tokens | Kimi K3$3 input / $15 output | LFM2.5-230M$0 input / $0 output | LFM2.5-230M has the lower combined listed price. |
| Generation speedtokens per second | Kimi K3Not available | LFM2.5-230MNot available | A complete speed comparison is not available. |
| First-answer latencyseconds to first token | Kimi K3Not available | LFM2.5-230MNot available | A complete latency comparison is not available. |
| Context windowmaximum listed tokens | Kimi K31.05M | LFM2.5-230M32K | Kimi K3 lists the larger context window. |
Benchmark Deep Dive
Agentic22 benchmarks
| Benchmark | Kimi K3 | LFM2.5-230M | Result |
|---|---|---|---|
| Terminal-Bench 2.0Source | 88.3% | — | Not comparable |
| BrowseCompSource | 91.2% | — | Not comparable |
| DeepSearchQASource | 95.0% | — | Not comparable |
| Toolathlon-VerifiedSource | 73.2% | — | Not comparable |
| MCP AtlasSource | 84.2% | — | Not comparable |
| AutomationBenchSource | 30.8% | — | Not comparable |
| JobBenchSource | 52.9% | — | Not comparable |
| APEX-AgentsSource | 37.6% | — | Not comparable |
| SpreadsheetBench 2Source | 34.8% | — | Not comparable |
| DECK-BenchSource | 73.5% | — | Not comparable |
| AA Agentic IndexSource | 50.1% | — | Not comparable |
| GDPval-AASource | 59.0% | — | Not comparable |
| GDPval-AASource | 1679 | — | Not comparable |
| AA BriefcaseSource | 1543 | — | Not comparable |
| AA AutomationBenchSource | 52.7% | — | Not comparable |
| AA EnterpriseOps-GymSource | 45.3% | — | Not comparable |
| AA Harvey LABSource | 94.6% | — | Not comparable |
| AA Tau3 BankingSource | 33.4% | — | Not comparable |
| aaTerminalBench21Source | 85% | — | Not comparable |
| APEX-Agents-AASource | 41.3% | — | Not comparable |
| AA ITBenchSource | 47.7% | — | Not comparable |
| BFCL v4Source | — | 21.0% | Not comparable |
Coding9 benchmarks
| Benchmark | Kimi K3 | LFM2.5-230M | Result |
|---|---|---|---|
| deepSweSource | 67.5% | — | Not comparable |
| FrontierSWESource | 81.2% | — | Not comparable |
| ProgramBenchSource | 77.8% | — | Not comparable |
| Kimi Code Bench v2Source | 72.9% | — | Not comparable |
| sweMarathonSource | 42% | — | Not comparable |
| PostTrain BenchSource | 36.6% | — | Not comparable |
| MLS-Bench LiteSource | 48.3% | — | Not comparable |
| AA Coding IndexSource | 76.2% | — | Not comparable |
| AA-SciCodeSource | 58.7% | — | Not comparable |
Reasoning2 benchmarks
KnowledgeKimi K3 wins11 benchmarks
| Benchmark | Kimi K3 | LFM2.5-230M | Result |
|---|---|---|---|
| GPQASource | 93.5% | 25.4% | Kimi K3 leads |
| GPQA-DSource | 93.5% | 25.4% | Kimi K3 leads |
| HLESource | 56% | — | Not comparable |
| HLE w/o toolsSource | 43.5% | — | Not comparable |
| Artificial Analysis Intelligence IndexSource | 57.1% | — | Not comparable |
| AA-GPQA DiamondSource | 93.5% | — | Not comparable |
| AA-HLESource | 44.3% | — | Not comparable |
| AA-Omniscience IndexSource | 18.4% | — | Not comparable |
| AA-Omniscience AccuracySource | 46.0% | — | Not comparable |
| AA-Omniscience Hallucination RateSource | 50.9% | — | Not comparable |
| MMLU-ProSource | — | 20.3% | Not comparable |
Multimodal15 benchmarks
| Benchmark | Kimi K3 | LFM2.5-230M | Result |
|---|---|---|---|
| OfficeQA ProSource | 63.3% | — | Not comparable |
| MMMU-ProSource | 81.6% | — | Not comparable |
| MMMU-Pro w/ PythonSource | 83.4% | — | Not comparable |
| CharXiv w/o toolsSource | 84.8% | — | Not comparable |
| CharXivSource | 91.3% | — | Not comparable |
| MathVisionSource | 94.3% | — | Not comparable |
| MathVision w/ PythonSource | 97.8% | — | Not comparable |
| BabyVision w/ PythonSource | 85.7% | — | Not comparable |
| ZeroBenchSource | 23.0% | — | Not comparable |
| ZeroBench w/ PythonSource | 41.0% | — | Not comparable |
| WorldVQA ForceAnswerSource | 51.0% | — | Not comparable |
| OmniDocBenchSource | 91.1% | — | Not comparable |
| PerceptionBenchSource | 58.5% | — | Not comparable |
| AA-MMMU-ProSource | 80.5% | — | Not comparable |
| Design Arena WebsiteSource | 1386 | — | Not comparable |
Frequently Asked Questions (2)
Which is better, Kimi K3 or LFM2.5-230M?
Kimi K3 and LFM2.5-230M are tied on the BenchAlign overall score, so the right pick depends on which category matters most for your use case.
Which is better for knowledge tasks, Kimi K3 or LFM2.5-230M?
Kimi K3 has the edge for knowledge tasks in this comparison, averaging 61 versus 21.2. Inside this category, GPQA is the benchmark that creates the most daylight between them.
Related Comparisons
Choose a model with this week’s evidence
Join 2,000+ readers for ranking moves, pricing changes, and the claims that still need proof.
One email each week. Unsubscribe anytime.