LFM2.5-24B-A2B released
LiquidAI · Model release
Source confirmed
Data as of October 13, 2026 · How the score is built
CurrentOpen WeightReasoning
Released Oct 13, 2026131K contextLiquid AI LFM2.5-24B-A2B documentation
LFM2.5-24B-A2B
Decision readingLFM2.5-24B-A2B scores 37.8 out of 100 and ranks #145 of 221. This profile shows 13 source-displayable benchmark rows; its strongest eligible category is Instruction Following at #61. API pricing is $0.1 input and $1 output per million tokens, with cached input at $0.05.
Released Oct 13, 2026 — see all recent releases
LFM2.5-24B-A2B will be repriced, updated or retired. Get each notice with its source and date. Follow model changes
Decision snapshot
Each value carries a field reference instead of floating alone. Markers compare this model with the current ranked and priced catalog; they are not absolute quality thresholds.
Capability
37.8/100
field median 49.8#145 of 221 ranked models
Public
#145of 221
Verified —
Price
$0.10input / $1 output
input median $0.75cached $0.050 · blended $0.55
Speed
Not measured
field median 90 tok/sTime to first token not measured
Context
131Ktokens
field median 256,000Maximum output length is tracked separately
Strongest published evidence
Instruction Following ranks #61. A well-rounded choice across a range of tasks.
Validate before choosing
13 published rows leave some tracked benchmark slots empty. Independent runtime speed has not been measured.
Source-linked · 13 displayable benchmark rows
Follow model changesCategory score record
Scores and ranks appear only where published evidence can be displayed. The table keeps the score, weight, cohort, and evidence state together.
| Category | Score | Rank | Percentile | Weight | Benchmarks | Evidence |
|---|---|---|---|---|---|---|
| AgenticRank #88 of 124Percentile 29thWeight 22%5 benchmarksVerified | 30.2 | 5 benchmarks | Verified | |||
| CodingRank #98 of 149Percentile 34thWeight 20%2 benchmarksVerified | 28.5 | 2 benchmarks | Verified | |||
| ReasoningWeight 17%0 benchmarksNot measured | Not measured | 0 benchmarks | Not measured | |||
| MultimodalWeight 12%0 benchmarksNot measured | Not measured | 0 benchmarks | Not measured | |||
| KnowledgeRank #112 of 181Percentile 38thWeight 12%3 benchmarksVerified | 40.8 | 3 benchmarks | Verified | |||
| MultilingualWeight 7%0 benchmarksNot measured | Not measured | 0 benchmarks | Not measured | |||
| Inst. FollowingRank #61 of 128Percentile 53rdWeight 5%2 benchmarksVerified | 75.6 | 2 benchmarks | Verified | |||
| MathWeight 5%1 benchmarkVerified | Score pending | 1 benchmark | Verified |
13 of 670 tracked benchmark slots have displayable evidence · bars run 0–100
Coverage detailsHow much of this is verified
Coverage is split by category so a strong number never hides a thin evidence base. Verified means the row is tied to a published source; provisional rows remain visible but separate.
- Agentic5/5 verified
- Coding2/2 verified
- ReasoningNot measured
- MultimodalNot measured
- Knowledge3/3 verified
- MultilingualNot measured
- Inst. Following2/2 verified
- Math1/1 verified
Verified sourceProvisionalNot measured
Capability shape
Each axis shows percentile within that category’s eligible cohort. The comparison outline is the median of the six nearest public-score peers; a collapsed vertex means the category is not rank-eligible.
LFM2.5-24B-A2B category percentile values
- Agentic29th percentile
- Coding34th percentile
- ReasoningNot eligible
- MultimodalNot eligible
- Knowledge38th percentile
- MultilingualNot eligible
- Instruction following53rd percentile
- MathNot eligible
The dashed outline is median of 6 nearest peers.
Eligible category ranks
- Agentic#88/124
- Coding#98/149
- ReasoningNot ranked
- MultimodalNot ranked
- Knowledge#112/181
- MultilingualNot ranked
- Inst. Following#61/128
- MathNot ranked
Top decileTop quartileMid-fieldNot eligible
Benchmark ledger
Coding opens by default. The marker compares each value with the best source-verified result in the catalog; provisional leaders do not set the reference. Expand the remaining categories for every published row.
Coding2 rows
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| SciCodeScientific Code Benchmark | Score32.7% | Versus best verified row Best verified: Sakana Fugu · 60.1% | Gap27.4 behind | Weight10% ref. weight | Provider exact |
| SWE-bench VerifiedSoftware Engineering Benchmark Verified | Score51% | Versus best verified row Best verified: Claude Opus 5 · 96% | Gap45 behind | WeightDisplay only | Provider exact |
Agentic5 rows
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| AutomationBench | Score27.7% | Versus best verified row Best verified: DeepSeek V4.1 Flash · 54.8% | Gap27.1 behind | Weight5% ref. weight | Provider exact |
| BFCL v4Berkeley Function Calling Leaderboard v4 | Score65.0% | Versus best verified row Best verified: BTL-3 · 88.5% | Gap23.5 behind | Weight3% ref. weight | Provider exact |
| τ³-bench resultsτ³-Bench Tool-Agent-User Evaluation | Score8.0% | Versus best verified row Best verified: Mercury 2.5 · 96.0% | Gap88 behind | Weight2% ref. weight | Provider exact |
| PinchBench | Score83.2% | Versus best verified row Best verified: Pokee-Isaac 28B · 95.7% | Gap12.5 behind | WeightDisplay only | Provider exact |
| WebArena-VerifiedWebArena-Verified Browser Agent Benchmark | Score26.8% | Versus best verified row Best verified: Muse Spark 1.1 · 69% | Gap42.3 behind | WeightDisplay only | Provider exact |
Knowledge3 rows
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| HLEHumanity's Last Exam | Score7.4% | Versus best verified row Best verified: Claude Fable 5.1 · 65% | Gap57.6 behind | Weight44% ref. weight | Provider exact |
| GPQAGraduate-Level Google-Proof Q&A | Score74.4% | Versus best verified row Best verified: GPT-6 Astra · 96% | Gap21.6 behind | Weight3% ref. weight | Provider exact |
| GPQA-DGPQA Diamond | Score74.4% | Versus best verified row Best verified: GPT-6 Astra · 96.0% | Gap21.6 behind | WeightDisplay only | Provider exact |
Inst. Following2 rows
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| IFBenchInstruction Following Benchmark | Score74.7% | Versus best verified row Best verified: MAI-Thinking-1 · 85% | Gap10.3 behind | WeightWeighted 70% | Provider exact |
| IFEvalInstruction-Following Eval | Score96.8% | Versus best verified row Best verified: LFM2.5-24B-A2B · 96.8% | GapBest verified | WeightDisplay only | Provider exact |
Math1 row
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| MATH-500MATH-500 Problem Set | Score96.0% | Versus best verified row Best verified: Ternary Bonsai 2 27B · 98.8% | Gap2.8 behind | WeightDisplay only | Provider exact |
Bars run 0–100; the dark tick marks the best source-verified value
All 13 rowsWhat it costs to get this score
Published API price against the public score. The x-axis uses a log scale; the dashed path marks models that are not beaten by a cheaper, higher-scoring option. Price uses average of published input and output rates.
Current modelExplore all models
LFM2.5-24B-A2B · 37.8 score · $0.55 blended per million tokens
The chart opens on the current model. Scroll horizontally to inspect the full price axis.
Horizontal: blended price per million tokens, log scale · Vertical: public score
Lineage
The sequence follows explicit supersedes links. Each score is estimated for that model; a relative can inform a sparse estimate but never sets a floor, so a newer release can score below an earlier one. Scores and prices remain blank when the corresponding public row or first-party rate is unavailable.
Reasoning
Radar
Full release historyLFM2.5-24B-A2B release history
Radar confirmed these at the source. Use LFM2.5-24B-A2B in your work? Explore Radar to follow supported changes and choose your alerts.
Radar
Spec sheet
Each documented value carries its source. Missing fields stay visible as not sourced or not published, rather than disappearing from the page.
- API model ID
- Not publishedLiquid AI launch post
- Context window
- 131KLiquid AI launch post
- Maximum output
- Not sourced yet
- Knowledge cutoff
- Not sourced yet
- Input modalities
- Not sourced yet
- Output modalities
- Not sourced yet
- Parameters
- Not sourced yet
- Availability
- Liquid AI publishes the open-weight 24B-parameter model with 2B active parameters and a 131,072-token context window. The launch post lists local inference formats and Liquid Console API access.
- Cloud regions
- Not tracked yet
- Lifecycle
- Current
- API capabilities
- Tool calling, structured outputs, and batch support are not tracked yet
- Prompt caching
- Published at $0.050 per million cached input tokensLiquid AI launch post
- Self-host
- Open weights available; hardware estimate not sourced
- Rate limits
- Not tracked yet
How to read this profile
The visual layer above carries the decisions. These notes preserve the model, ranking, coverage, and family context behind the numbers.
LFM2.5-24B-A2B ranks #145 of 221 on the public leaderboard with a score of 37.8/100. It does not yet have enough sourced coverage for a verified position.
LFM2.5-24B-A2B is a open weight model with a 131K context window. It uses an explicit reasoning mode, which can improve complex problem solving while adding latency and token use.
Liquid AI publishes the open-weight 24B-parameter model with 2B active parameters and a 131,072-token context window. The launch post lists local inference formats and Liquid Console API access.
Official exact-value snapshot from Liquid AI's October 13, 2026 launch post. BenchLM maps the directly named GPQA Diamond, HLE, SWE-Bench Verified (OpenCode), SciCode, MATH-500, IFEval, IFBench, AutomationBench, BFCLv4, tau3-Bench Banking, PinchBench, and WebArena Verified results. ToolSandbox, LiveClaw, and Multi-IF do not have exact compatible catalog keys. The launch measurements are provider-run and remain identified as such in row-level source verification.
Its explicit predecessor is LFM2-24B-A2B. 13 of 670 tracked benchmark slots currently have displayable evidence. Missing categories stay blank.
Its strongest eligible category is Instruction Following at #61, while its lowest eligible position is Knowledge at #112. a well-rounded choice across a range of tasks.
Last updated October 13, 2026. Runtime fields remain blank until a sourced snapshot exists.
Deployment options
Self-host and provider-specific paths stay separate from benchmark evidence so operating constraints are visible before a score becomes the whole decision.
Published weights are available, but BenchLM does not yet have a sourced parameter and VRAM profile for this exact model. Hardware cost estimates stay unavailable until that sizing record is complete.
Estimate VRAM from known parametersQuestions
How does LFM2.5-24B-A2B perform overall in AI benchmarks?
LFM2.5-24B-A2B ranks #145 out of 221 models on the public BenchAlign leaderboard, with a score of 37.8/100. Its evidence status is Estimated, and this profile shows 13 source-displayable benchmark rows. The label describes evidence depth, not a provider quality claim; inspect category rows before choosing a workload.
Is LFM2.5-24B-A2B good for knowledge and understanding?
LFM2.5-24B-A2B ranks #112 out of 181 eligible models for knowledge and understanding, with a public category score of 40.8/100. Higher-ranked alternatives are available for workloads where this category decides the choice. Check the underlying rows before treating the aggregate as a workload guarantee.
Is LFM2.5-24B-A2B good for coding and programming?
LFM2.5-24B-A2B ranks #98 out of 149 eligible models for coding and programming, with a public category score of 28.5/100. Higher-ranked alternatives are available for workloads where this category decides the choice. Check the underlying rows before treating the aggregate as a workload guarantee.
Is LFM2.5-24B-A2B good for mathematics?
LFM2.5-24B-A2B has source-displayable benchmark coverage for mathematics, but the public category table does not assign it a rank there. The individual rows remain available for inspection. A missing category position means the evidence threshold was not met; it does not convert the model's unmeasured work into a zero.
Is LFM2.5-24B-A2B good for agentic tool use and computer tasks?
LFM2.5-24B-A2B ranks #88 out of 124 eligible models for agentic tool use and computer tasks, with a public category score of 30.2/100. Higher-ranked alternatives are available for workloads where this category decides the choice. Check the underlying rows before treating the aggregate as a workload guarantee.
Is LFM2.5-24B-A2B good for instruction following?
LFM2.5-24B-A2B ranks #61 out of 128 eligible models for instruction following, with a public category score of 75.6/100. Higher-ranked alternatives are available for workloads where this category decides the choice. Check the underlying rows before treating the aggregate as a workload guarantee.
Is LFM2.5-24B-A2B open source?
LFM2.5-24B-A2B is an open-weight model from LiquidAI. Its weights can be downloaded for local or hosted deployment, subject to the published license. Open weight does not automatically mean open source: training data and training code may remain private, and commercial restrictions can still apply.
Does LFM2.5-24B-A2B have full benchmark coverage on BenchLM?
No. LFM2.5-24B-A2B currently has 14 source-displayable rows across 670 tracked benchmark slots. The profile exposes published, non-generated evidence and leaves missing categories blank until an exact evaluation is available. Coverage describes how much was measured; it is not a penalty added to an individual benchmark result.
What is the context window size of LFM2.5-24B-A2B?
LFM2.5-24B-A2B has a documented context window of 131K. That figure is the maximum combined prompt and retained-conversation space reported for this exact model; it is not the maximum output length. The profile keeps output limits separate because providers often publish those limits independently.