Skip to content
Model Pareto

Motif 3

Motif TechnologiesOpen weightsartificialanalysis.ai ↗
Input
—
Output
—
Speed
~92tok/sest.
Context
262Ktokens

Long-tail import from Artificial Analysis (2026-09-24); AA variant: Motif 3. 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
4× B200 180GB
FP8 weights
Throughput
~19,600 tok/s
many requests batched
Price
$0.13–1.19 /M tok
busy → light use
On your own machine
Mac Studio M5 Ultra 512GB (int4, ~130 tok/s single-stream)
single consumer GPU or Mac
Assumptions (5)
  • FP8 weights (314B params, 13.2B active per token) + 25% KV-cache headroom ≈ 393 GB
  • 4× B200 180GB at $5.98–$14.24/GPU-hour on-demand (2026-09-24)
  • ~19,600 output tok/s aggregate at batch 561 (bandwidth-bound); MoE compute scales with active params
  • Blended 3:1 input:output; prefill ~322,159 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.

Coding

~#14 of 21 (estimated)~72.8%
  1. #1Claude Opus 596.0%
  2. #2Claude Sonnet 585.2%
  3. #13Claude Haiku 4.573.3%
  4. ~#14Motif 3Estimated72.8%
  5. #14Nemotron 3.5 Super VL71.1%
No published score — estimated from related benchmarks
DeepSWE v1.1Estimated
~#39 of 46 (estimated)~43.4%
  1. #1Gemini 4 Argon78.8%
  2. #2DeepSeek V4.1 Flash 101777.0%
  3. #38Ripple-65B Preview44.3%
  4. ~#39Motif 3Estimated43.4%
  5. #39Beam 1.0 501B A23B43.3%
No published score — estimated from related benchmarks
~#34 of 102 (estimated)~14.5%
  1. #1Kimi K3.1 Pro67.2%
  2. #2Claude Sonnet 5.563.6%
  3. #33GPT-5.514.6%
  4. ~#34Motif 3Estimated14.5%
  5. #34Claude Sonnet 514.1%
No published score — estimated from related benchmarks
LiveCodeBenchEstimated
~#15 of 28 (estimated)~70.9%
  1. #1Qwen3.8-Omni-Flash92.6%
  2. #2Qwen3.8 Flash-Next91.9%
  3. #14Maple-Preview75.1%
  4. ~#15Motif 3Estimated70.9%
  5. #15Mistral Small 464.0%
No published score — estimated from related benchmarks
~#23 of 26 (estimated)~35.1
  1. #1Claude Opus 5.561.7
  2. #2Claude Opus 559.7
  3. #22Claude Haiku 5.537.3
  4. ~#23Motif 3Estimated35.1
  5. #23Gemini 3.8 Flash33.3
No published score — estimated from related benchmarks

Hard reasoning

Agents

~#12 of 120 (estimated)~91.9%
  1. #1Step 3.7 Flash98.5%
  2. #2DeepSeek-V4-Pro-0813-nvfp4-DSpark98.3%
  3. #11Ring-2.6-1T92.4%
  4. ~#12Motif 3Estimated91.9%
  5. #12MiMo-V2.590.6%
No published score — estimated from related benchmarks
~#9 of 15 (estimated)~76.2%
  1. #1Kimi K384.8%
  2. #2Qwen3.8 27B84.3%
  3. #8GPT-5.578.7%
  4. ~#9Motif 3Estimated76.2%
  5. #9Gemini 3.5 Flash-Lite74.0%
No published score — estimated from related benchmarks
OSWorld 2.0Estimated
~last of 5 (estimated)below measured range (<47.5%)
  1. #1GLM-5.3 Flash59.1%
  2. #2Kimi K358.3%
  3. #4GPT-5.547.5%
  4. ~#5Motif 3Estimated<47.5%
No published score — estimated from related benchmarks
BrowseCompEstimated
~#14 of 22 (estimated)~82.8%
  1. #1Atria Dawn Preview92.5%
  2. #2GPT-6 Astra91.5%
  3. #13Nex-N2.5-Mini83.4%
  4. ~#14Motif 3Estimated82.8%
  5. #14Nemotron 3.5 Super VL78.0%
No published score — estimated from related benchmarks
~#21 of 92 (estimated)~1501 Elo
  1. #1Claude Opus 5.51866
  2. #2Muse Wonder Full1857
  3. #20Kimi K31533
  4. ~#21Motif 3Estimated1501
  5. #21GPT-6 Sol1487
No published score — estimated from related benchmarks
~#29 of 85 (estimated)~39.5%
  1. #1Claude Fable 5.276.4%
  2. #2Muse Wonder Full71.3%
  3. #28GPT-5.547.3%
  4. ~#29Motif 3Estimated39.5%
  5. #29K2 Horizon 375B A23B37.2%
No published score — estimated from related benchmarks