- LiveClaw (OpenClaw agent, assumed to be LiveClawBench) 45.6
- Claw-Eval General (EN), average 68.6
- Claw-Eval General (EN), pass^3 51.4
- BrowseComp-Plus (OpenClaw subset) 37
- WebArena-Verified 38.4
- EnterpriseOps-Gym 28.2
- AutomationBench (public) 41.3
- BFCL V4 71.4
- ToolSandbox 80.4
- MCP-Atlas 32.6
- τ³-bench 68
- SWE-bench Verified (OpenCode) 43.4
- SWE-bench Multilingual (OpenCode) 23
- Terminal-Bench 2.1 (OpenCode) 12.9
- LiveCodeBench v6 73.5
- LiveCodeBench Pro 25Q2 31.7
- SciCode 28.5
- IFEval 96.7
- IFBench 73
- IFStruct 93.7
- Multi-IF 89.1
- GPQA Diamond 63.4
- Best of Humanity's Last Exam no tools 7.17 / with tools 13.90 13.9
- AA-Omniscience (public) index 0.50, accuracy 7.07, non-hallucination 92.93 0.5
- AA-LCR 52.8
- RULER 128K 65.6
LFM3-1.7B
LFM3-1.7B ranks first on Multi-IF.
25
Benchmarks with a score
1
Ranked first, out of five or more models
2
In the top three
Liquid AI Oct 2026
LFM3-1.7B
Open weights Flagship
132.8
#252 of 404
Self-Reported Index, from the lab's own numbers. No Epoch Capabilities Index score yet.
- IFBench 73% #30 /80
- SciCode 28.5% #44 /49
- AA-LCR (Artificial Analysis Long Context Reasoning) 52.8% #30 /38
Where it ranks
Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.
Knowledge and language
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| Multi-IF | 89.1% | #1 of 28 | Lab model card | |
| IFEval | 96.7% | #6 of 135 | Lab model card | |
| IFBench | 73% | #30 of 80 | Lab model card | |
| IFStruct | 93.7% | #1 of 2 | Lab model card | |
| AA-Omniscience | 0.5 | #6 of 10 | Lab model card | |
| AA-LCR (Artificial Analysis Long Context Reasoning) | 52.8% | #30 of 38 | Lab model card | |
| Humanity's Last Exam | 13.9% | #126 of 154 | Lab model card | |
| RULER | 65.6% | #40 of 42 | Lab model card |
Math and science
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| GPQA Diamond | 63.4% | #212 of 302 | Lab model card |
Coding
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| LiveCodeBench v6 | 73.5% | #50 of 90 | Lab model card | |
| SWE-bench Verified | 43.4% | #139 of 155 | Lab model card | |
| SciCode | 28.5% | #44 of 49 | Lab model card | |
| SWE-bench Multilingual | 23% | #73 of 73 | Lab model card |
Agents and tools
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| BFCL V4 | 71.4% | #7 of 30 | Lab model card | |
| ToolSandbox | 80.4% | #2 of 5 | Lab model card | |
| AutomationBench | 41.3% | #22 of 47 | Lab model card | |
| tau3-bench | 68% | #7 of 14 | Lab model card | |
| Claw-Eval | 68.6% | #8 of 13 | Lab model card | |
| BrowseComp-Plus | 37% | #3 of 4 | Lab model card | |
| Claw-Eval Pass^3 | 51.4% | #11 of 13 | Lab model card | |
| WebArena-Verified | 38.4% | #6 of 7 | Lab model card | |
| MCP Atlas | 32.6% | #45 of 46 | Lab model card | |
| Terminal-Bench 2.1 | 12.9% | #56 of 56 | Lab model card | |
| EnterpriseOps-Gym | 28.2% | #3 of 3 | Lab model card | |
| LiveClawBench | 45.6% | #2 of 2 | Lab model card |
Its lab's numbers put it at 132.8, number 252 of 404
The Self-Reported Index fits one score to the 18 capability results Liquid AI reported for this model, on the scale of the Epoch Capabilities Index. Labs choose which benchmarks to show and how to run them, so read it as the lab's claim, not an independent measurement. The 90% interval runs from 129.0 to 136.7. How we compute it.
- IFEval 96.7
- Multi-IF 89.1
- LiveCodeBench v6 73.5
- IFBench 73
- BFCL V4 71.4
- Claw-Eval 68.6
- tau3-bench 68
- RULER 65.6
- GPQA Diamond 63.4
- AA-LCR (Artificial Analysis Long Context Reasoning) 52.8
- Claw-Eval Pass^3 51.4
- SWE-bench Verified 43.4
- AutomationBench 41.3
- MCP Atlas 32.6
- SciCode 28.5
- SWE-bench Multilingual 23
- Humanity's Last Exam 13.9
- Terminal-Bench 2.1 12.9
Its launch post
The benchmarks the lab chose to headline, with the scores it gave.