Inference
EN
LiquidAI

LFM3-1.7B

LFM3-1.7B is a reasoning model from LiquidAI in the LFM3 family. 16 benchmarks count toward its index, in 5 categories.

  • Open weights
  • Reasoning
  • Input: text
  • Released 2026-10
RankedEnough results for an index.
IndexAll categories, out of 100.41.5±6.5Median 51.0
CoverageIndex weight that has results.75%
SpeedOutput tokens per second.218/sMedian 67/s
First tokenSeconds to the first output token.0.3 sMedian 1.3 s
Input / 1MUS dollars per 1M input tokens.$0.08Median $0.60
Output / 1MUS dollars per 1M output tokens.$0.20Median $2.55
ContextMaximum tokens in one request.262KMedian 262K
EloLMArena rating and rank.—Median 1422

Details

From OpenRouter
API IDThe model ID on OpenRouter.
liquid/lfm-3-1.7b
Max outputMaximum output tokens per request.
8K
Knowledge cutoffLast date of training data.
—
ToolsTool calls on OpenRouter.
Yes
JSON outputOutput follows a given JSON schema.
Yes
Cache read / 1MUS dollars per 1M cached tokens.
$0.02
WeightsLink to the published weights.
Hugging Face (opens in a new tab) ↗
RetiresDate OpenRouter removes the model.
—

Run locally

1.73B parameters
16-bit
4.6 GB
8-bit
2.9 GB
4-bit
2.1 GB
Weights
Hugging Face (opens in a new tab) ↗

Memory need with a short context. The 4-bit copy fits in 8 GB. Compare local models

Capabilities

Index per category, out of 100.
Out of 100
AgenticMulti-step tasks with tools.7 counted46.4158 / 218
CodingCode writing and repair.3 counted37.0187 / 209
ReasoningLogic problems and puzzles.1 counted43.5—
MultimodalTasks with images and text.None counted——
KnowledgeFacts and expert knowledge.3 counted28.8242 / 268
MultilingualTasks in many languages.None counted——
InstructionTasks with strict prompt rules.2 counted57.619 / 63
MathMath problems.None counted——

Results

16 counted
BenchmarkThe benchmark name.ResultThe score from the source.PlacePlace among results on this benchmark.
Instruction-Following EvalInstruction96.7%2 / 70
τ³-Bench Tool-Agent-User EvaluationAgentic68.0%9 / 32
Berkeley Function Calling Leaderboard v4Agentic71.4%8 / 26
Claw-EvalAgenticAbove fitFar above this model's index.68.6%13 / 42
AutomationBenchAgenticAbove fitFar above this model's index.41.3%19 / 32
LiveCodeBench v6Coding73.5%37 / 59
Instruction Following BenchmarkInstruction73.0%36 / 57
Artificial Analysis Long Context ReasoningReasoning52.8%150 / 217
Graduate-Level Google-Proof Q&AKnowledge63.4%107 / 141
GPQA Diamond · BenchLMKnowledge63.4%76 / 84
Scientific Code BenchmarkCoding28.5%32 / 35
Software Engineering Benchmark VerifiedCoding43.4%106 / 114
MCP AtlasAgentic32.6%50 / 52
Humanity's Last ExamKnowledge7.2%76 / 77
Terminal-Bench 2.1 · BenchLMAgentic12.9%85 / 86
Humanity's Last Exam with toolsAgentic13.9%36 / 36
Artificial Analysis Omniscience AccuracyKnowledge7.1%164 / 164
IFStructInstruction93.7%—
Multi-IFInstruction89.1%—
ToolSandboxAgentic80.4%—
RULER at 128K contextReasoning65.6%—
LiveClawAgentic45.6%—
WebArena-Verified Browser Agent BenchmarkAgentic38.4%—
LiveCodeBench ProCoding31.7%—
EnterpriseOps-Gym (provider runs)Agentic28.2%—

Sources

BenchLMCC BY-NC 4.0OpenRouterTerms restrict reuse

History

5 releases
Index41.9 → 41.5
29 Oct 202612 Nov 2026
Output / 1M$0.20 → $0.20
11 Nov 202612 Nov 2026

Questions

How much does LFM3-1.7B cost?
$0.08 per 1M input tokens, $0.20 per 1M output tokens. A free tier is also available.
How good is LFM3-1.7B?
Its index is 41.5 out of 100. That is place 203 of 283 ranked models.
What is LFM3-1.7B best at?
Instruction: 57.6 out of 100, place 19 of 63.
How fast is LFM3-1.7B?
218 output tokens per second, 0.3 s to the first token.
What is the context window of LFM3-1.7B?
262K tokens.
Is LFM3-1.7B open source?
Yes. Its weights are published.
When was LFM3-1.7B released?
October 2026.

Badge

LFM3-1.7B on Inference360: place 203 of 283, index 41.5
HTML
<a href="https://inference360.ai/models/lfm3-1-7b"><img src="https://inference360.ai/badges/lfm3-1-7b.svg" alt="LFM3-1.7B on Inference360: place 203 of 283, index 41.5" height="20"></a>
Markdown
[![LFM3-1.7B on Inference360: place 203 of 283, index 41.5](https://inference360.ai/badges/lfm3-1-7b.svg)](https://inference360.ai/models/lfm3-1-7b)

Compare with

Claude 3.7 Sonnet Thinking41.4Granite 4.2 30B41.6GPT-OSS 20B41.7Grok Code Fast 141.2LFM2.5-24B-A2B39.8

Lower price

dots3-note Preview63.4 · FreeApodex 1.1 Mini55.7 · FreeNemotron 3 Ultra50.8 · Free

More from LiquidAI

All models
d1-2BUnrankedd1-Omni-2BUnrankedLFM3-1.7B-BaseUnrankedLFM3-Audio-2BUnrankedLFM3-Omni-2BUnranked