Skip to content
Model Pareto

Molmo2-8B

Ai2 (Allen Institute)Open weightsartificialanalysis.ai ↗
Input
—
Output
—
Speed
~97tok/sest.
Context
37Ktokens

Long-tail import from Artificial Analysis (2026-09-24); AA variant: Molmo2-8B. 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
~10,294 tok/s
many requests batched
Price
$0.081–0.72 /M tok
busy → light use
On your own machine
RTX 5090 32GB (int4, ~290 tok/s single-stream)
single consumer GPU or Mac
Assumptions (5)
  • BF16 weights (8.66B params) + 25% KV-cache headroom ≈ 22 GB
  • 1× B200 180GB at $5.98–$14.24/GPU-hour on-demand (2026-09-24)
  • ~10,294 output tok/s aggregate at batch 256 (bandwidth-bound)
  • Blended 3:1 input:output; prefill ~61,381 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. ~#2Molmo2-8BEstimated71.3%
  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. ~#12Molmo2-8BEstimated<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. ~#8Molmo2-8BEstimated<75.8%
No published score — estimated from related benchmarks
GDP.pdf (AA)Estimated
~#46 of 48 (estimated)~0.1%
  1. #1Muse Wonder Full32.6%
  2. #2GPT-6 Astra31.0%
  3. #45Ministral 3 8B0.8%
  4. ~#46Molmo2-8BEstimated0.1%
  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. ~#25Molmo2-8BEstimated<9.0%
No published score — estimated from related benchmarks

Hard reasoning

Agents

~#13 of 15 (estimated)~51.4%
  1. #1Kimi K384.8%
  2. #2Qwen3.8 27B84.3%
  3. #12Nemotron 3.5 Super VL NVFP452.3%
  4. ~#13Molmo2-8BEstimated51.4%
  5. #13Nemotron 3.5 Super VL51.1%
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. ~#3Molmo2-8BEstimated53.8%
  4. #3GPT-5.6 Terra50.2%
No published score — estimated from related benchmarks
BrowseCompEstimated
~#21 of 22 (estimated)~35.8%
  1. #1Atria Dawn Preview92.5%
  2. #2GPT-6 Astra91.5%
  3. #20Nemotron 3 Ultra44.4%
  4. ~#21Molmo2-8BEstimated35.8%
  5. #21Mistral Small 421.3%
No published score — estimated from related benchmarks
~#64 of 92 (estimated)~584 Elo
  1. #1Claude Opus 5.51866
  2. #2Muse Wonder Full1857
  3. #63gpt-oss-120b596
  4. ~#64Molmo2-8BEstimated584
  5. #64Granite 4.2 30B564
No published score — estimated from related benchmarks
~#85 of 85 (estimated)~0.2%
  1. #1Claude Fable 5.276.4%
  2. #2Muse Wonder Full71.3%
  3. #79Magistral Small 1.20.2%
  4. ~#85Molmo2-8BEstimated0.2%
No published score — estimated from related benchmarks
~last of 91 (estimated)below measured range (<3.3%)
  1. #1Muse Spark 1.350.5%
  2. #2GLM-5.350.3%
  3. #90Llama 4 Scout3.3%
  4. ~#91Molmo2-8BEstimated<3.3%
No published score — estimated from related benchmarks