Skip to content
Model Pareto

MiniCPM-V 4.6 1.3B

OpenBMB (MiniCPM)Open weightsartificialanalysis.ai ↗
Input
—
Output
—
Speed
~120tok/sest.
Context
262Ktokens

Long-tail import from Artificial Analysis (2026-09-24); AA variant: MiniCPM-V 4.6 1.3B. Price = AA median host price.

Rankings

Self-hostingEstimated

How we estimate →

What it would cost to run these open weights yourself on rented GPUs. No API sells this model, so this is its price on the chart.

Hardware
1× H200 141GB
BF16 weights
Throughput
~17,912 tok/s
many requests batched
Price
$0.024–0.20 /M tok
busy → light use
On your own machine
RTX 5090 32GB (int4, 300+ tok/s single-stream)
single consumer GPU or Mac
Assumptions (5)
  • BF16 weights (1.3B params) + 25% KV-cache headroom ≈ 3 GB
  • 1× H200 141GB at $3.59–$7.91/GPU-hour on-demand (2026-09-24)
  • ~17,912 output tok/s aggregate at batch 256 (bandwidth-bound)
  • Blended 3:1 input:output; prefill ~179,732 tok/s
  • Low = 75% utilization at the low GPU price; high = 20% at the high price

Benchmark scores

~ italic, dashed = no published score yet, estimated from related benchmarks.

Writing & chat

Vision

MMMUEstimated
~#2 of 3 (estimated)~71.3%
  1. #1Claude Haiku 4.573.2%
  2. ~#2MiniCPM-V 4.6 1.3BEstimated71.3%
  3. #2Llama 4 Maverick69.4%
No published score — estimated from related benchmarks
~last of 12 (estimated)below measured range (<74.5%)
  1. #1Gemini 3.8 Flash86.2%
  2. #2Kimi K384.8%
  3. #11Gemini 3.5 Flash-Lite74.5%
  4. ~#12MiniCPM-V 4.6 1.3BEstimated<74.5%
No published score — estimated from related benchmarks
~last of 8 (estimated)below measured range (<75.8%)
  1. #1MiniMax-M391.6%
  2. #2Kimi K391.1%
  3. #7Muse Glimmer75.8%
  4. ~#8MiniCPM-V 4.6 1.3BEstimated<75.8%
No published score — estimated from related benchmarks
GDP.pdf (AA)Estimated
~#46 of 48 (estimated)~0.2%
  1. #1Muse Wonder Full32.6%
  2. #2GPT-6 Astra31.0%
  3. #45Ministral 3 8B0.8%
  4. ~#46MiniCPM-V 4.6 1.3BEstimated0.2%
  5. #46Ministral 3 3B0.0%
No published score — estimated from related benchmarks
ChartographyEstimated
~last of 25 (estimated)below measured range (<9.0%)
  1. #1Claude Fable 5.273.4%
  2. #2GPT-6 Astra71.0%
  3. #24Mistral Large 39.0%
  4. ~#25MiniCPM-V 4.6 1.3BEstimated<9.0%
No published score — estimated from related benchmarks

Hard reasoning

Agents

~#12 of 15 (estimated)~61.8%
  1. #1Kimi K384.8%
  2. #2Qwen3.8 27B84.3%
  3. #11Muse Glimmer65.9%
  4. ~#12MiniCPM-V 4.6 1.3BEstimated61.8%
  5. #12Nemotron 3.5 Super VL NVFP452.3%
No published score — estimated from related benchmarks
OSWorld 2.0Estimated
~#3 of 5 (estimated)~53.8%
  1. #1GLM-5.3 Flash59.1%
  2. #2Kimi K358.3%
  3. ~#3MiniCPM-V 4.6 1.3BEstimated53.8%
  4. #3GPT-5.6 Terra50.2%
No published score — estimated from related benchmarks
~#47 of 92 (estimated)~900 Elo
  1. #1Claude Opus 5.51866
  2. #2Muse Wonder Full1857
  3. #46GPT-5.4 nano937
  4. ~#47MiniCPM-V 4.6 1.3BEstimated900
  5. #47Qwen3.6 35B A3B879
No published score — estimated from related benchmarks
~#78 of 91 (estimated)~5.7%
  1. #1Muse Spark 1.350.5%
  2. #2GLM-5.350.3%
  3. #76Trinity Large Thinking5.8%
  4. ~#78MiniCPM-V 4.6 1.3BEstimated5.7%
  5. #78Granite 4.2 3B5.6%
No published score — estimated from related benchmarks