tinyanalysis

Compare / LFM2.5 1.2B Thinking vs MiniCPM-V 4.6

LFM2.5 1.2B Thinking vs MiniCPM-V 4.6

Updated November 2026

Verdict

LFM2.5 1.2B Thinking scores higher on the Small LM Index: 34.7 vs 27.0, although their likely ranges overlap, so the lead is not certain. On the 5 benchmarks both labs report, LFM2.5 1.2B Thinking wins 3 and MiniCPM-V 4.6 wins 2. Both need about 1.4 GB of memory at 4-bit quantization. Neither model has a hosted API on OpenRouter, so you would run both yourself.

Pick LFM2.5 1.2B Thinking if…

  • you want the strongest model overall
  • you need precise instruction following

Pick MiniCPM-V 4.6 if…

  • you build agents that call tools
  • you work with long documents
  • you need a permissive license (Apache-2.0)
  • you need image input
  • you want the more widely used model (more quantized builds and community support)

Specifications

LFM2.5 1.2B ThinkingMiniCPM-V 4.6
Small LM Index34.7 (likely 32–38)27.0 (likely 19–39)
LabLiquid AI · United StatesOpenBMB (MiniCPM) · China
Released2026-01-202026-04-13
Parameters1.2B (dense)1.3B (dense)
Memory at 4-bit~1.3 GB~1.4 GB
Context window32K tokens262K tokens
Inputtexttext, image, video
ReasoningAlways thinksNo thinking
LicenseLFM Open LicenseApache-2.0
API price / 1M tokensSelf-host onlySelf-host only
Hugging Face downloads (30 days)8.3K196.5K

Benchmarks

Higher is better. Bold marks the winner where both labs report a score. ≈ marks the index model's estimate for a score that wasn't reported, ✓ an independent measurement and * a score measured by another lab.

BenchmarkLFM2.5 1.2B ThinkingMiniCPM-V 4.6
MMLU-Pro Knowledge49.7≈ 50
GPQA Diamond Knowledge33.9✓30.5✓
Humanity's Last Exam Reasoning6.2✓5.1✓
AIME Math31.7≈ 17
τ²-Bench Agentic19.6✓87.7✓
IFBench Instruction following41.8✓26.7✓
IFEval Instruction following88.4—
BFCL Agentic57.0—
MATH-500 Math88.0—
AA-LCR Long context0.0✓7.0✓

Scores are mostly reported by each lab with its own settings, so small gaps are noise. See the methodology. Open in the interactive comparison to add more models.

Questions

Which is better, LFM2.5 1.2B Thinking or MiniCPM-V 4.6?

LFM2.5 1.2B Thinking scores higher on the Small LM Index: 34.7 vs 27.0, although their likely ranges overlap, so the lead is not certain. On the 5 benchmarks both labs report, LFM2.5 1.2B Thinking wins 3 and MiniCPM-V 4.6 wins 2.

Which is cheaper to run?

Neither model has a hosted API on OpenRouter, so you would run both yourself. Both need about 1.4 GB of memory at 4-bit quantization.

Which has the longer context window?

MiniCPM-V 4.6 has the longer context window: 262K tokens against 32K.

Can I use LFM2.5 1.2B Thinking and MiniCPM-V 4.6 commercially?

LFM2.5 1.2B Thinking is released under LFM Open License and MiniCPM-V 4.6 under Apache-2.0. Apache 2.0 and MIT allow commercial use; other licenses can add conditions, so read the license before shipping a product.

Related comparisons