Skip to content
Model Pareto

Step3 VL 10B

StepFunOpen weightsartificialanalysis.ai ↗
Input
—
Output
—
Speed
~100tok/sest.
Context
66Ktokens

Long-tail import from Artificial Analysis (2026-09-24); AA variant: Step3 VL 10B. 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× B200 180GB
BF16 weights
Throughput
~9,402 tok/s
many requests batched
Price
$0.091–0.81 /M tok
busy → light use
On your own machine
RTX 5090 32GB (int4, ~250 tok/s single-stream)
single consumer GPU or Mac
Assumptions (5)
  • BF16 weights (10.17B params) + 25% KV-cache headroom ≈ 25 GB
  • 1× B200 180GB at $5.98–$14.24/GPU-hour on-demand (2026-09-24)
  • ~9,402 output tok/s aggregate at batch 256 (bandwidth-bound)
  • Blended 3:1 input:output; prefill ~52,268 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. ~#2Step3 VL 10BEstimated71.3%
  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. ~#12Step3 VL 10BEstimated<74.5%
No published score — estimated from related benchmarks
~#6 of 8 (estimated)~79.2%
  1. #1MiniMax-M391.6%
  2. #2Kimi K391.1%
  3. #5Claude Haiku 4.579.6%
  4. ~#6Step3 VL 10BEstimated79.2%
  5. #6Mistral Small 476.4%
No published score — estimated from related benchmarks
GDP.pdf (AA)Estimated
~#37 of 48 (estimated)~6.7%
  1. #1Muse Wonder Full32.6%
  2. #2GPT-6 Astra31.0%
  3. #36GPT-5.4 nano7.8%
  4. ~#37Step3 VL 10BEstimated6.7%
  5. #37Qwen3.6 35B A3B6.6%
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. ~#25Step3 VL 10BEstimated<9.0%
No published score — estimated from related benchmarks

Hard reasoning

Agents

~#12 of 15 (estimated)~56.7%
  1. #1Kimi K384.8%
  2. #2Qwen3.8 27B84.3%
  3. #11Muse Glimmer65.9%
  4. ~#12Step3 VL 10BEstimated56.7%
  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. ~#3Step3 VL 10BEstimated53.8%
  4. #3GPT-5.6 Terra50.2%
No published score — estimated from related benchmarks
BrowseCompEstimated
~#21 of 22 (estimated)~41.1%
  1. #1Atria Dawn Preview92.5%
  2. #2GPT-6 Astra91.5%
  3. #20Nemotron 3 Ultra44.4%
  4. ~#21Step3 VL 10BEstimated41.1%
  5. #21Mistral Small 421.3%
No published score — estimated from related benchmarks
~#38 of 92 (estimated)~1052 Elo
  1. #1Claude Opus 5.51866
  2. #2Muse Wonder Full1857
  3. #37Inkling1076
  4. ~#38Step3 VL 10BEstimated1052
  5. #38Grok Build 0.1 06161052
No published score — estimated from related benchmarks
~#47 of 85 (estimated)~10.8%
  1. #1Claude Fable 5.276.4%
  2. #2Muse Wonder Full71.3%
  3. #46K2 Horizon 3.7B14.7%
  4. ~#47Step3 VL 10BEstimated10.8%
  5. #47Qwen3.5 397B A17B6.9%
No published score — estimated from related benchmarks
~#57 of 91 (estimated)~12.6%
  1. #1Muse Spark 1.350.5%
  2. #2GLM-5.350.3%
  3. #56gpt-oss-120b12.8%
  4. ~#57Step3 VL 10BEstimated12.6%
  5. #57Gemma 4 26B A4B12.0%
No published score — estimated from related benchmarks