Skip to content
Model Pareto

EXAONE 4.0 32B

LG AI Research (EXAONE)Open weightsartificialanalysis.ai ↗
Input
—
Output
—
Speed
~83tok/sest.
Context
131Ktokens

Long-tail import from Artificial Analysis (2026-09-24); AA variant: EXAONE 4.0 32B (Non-reasoning). 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
FP8 weights
Throughput
~4,898 tok/s
many requests batched
Price
$0.16–1.46 /M tok
busy → light use
On your own machine
RTX 5090 32GB (int4, ~78 tok/s single-stream)
single consumer GPU or Mac
Assumptions (5)
  • FP8 weights (32B params) + 25% KV-cache headroom ≈ 40 GB
  • 1× B200 180GB at $5.98–$14.24/GPU-hour on-demand (2026-09-24)
  • ~4,898 output tok/s aggregate at batch 140 (bandwidth-bound)
  • Blended 3:1 input:output; prefill ~33,223 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

Hard reasoning

Agents

~#12 of 15 (estimated)~52.7%
  1. #1Kimi K384.8%
  2. #2Qwen3.8 27B84.3%
  3. #11Muse Glimmer65.9%
  4. ~#12EXAONE 4.0 32BEstimated52.7%
  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. ~#3EXAONE 4.0 32BEstimated53.8%
  4. #3GPT-5.6 Terra50.2%
No published score — estimated from related benchmarks
BrowseCompEstimated
~#21 of 22 (estimated)~35.2%
  1. #1Atria Dawn Preview92.5%
  2. #2GPT-6 Astra91.5%
  3. #20Nemotron 3 Ultra44.4%
  4. ~#21EXAONE 4.0 32BEstimated35.2%
  5. #21Mistral Small 421.3%
No published score — estimated from related benchmarks
~#61 of 92 (estimated)~660 Elo
  1. #1Claude Opus 5.51866
  2. #2Muse Wonder Full1857
  3. #60MiniCPM5-2B694
  4. ~#61EXAONE 4.0 32BEstimated660
  5. #61Nemotron 3.5 Lightning624
No published score — estimated from related benchmarks
~#79 of 85 (estimated)~0.2%
  1. #1Claude Fable 5.276.4%
  2. #2Muse Wonder Full71.3%
  3. #77Granite 4.2 3B0.4%
  4. ~#79EXAONE 4.0 32BEstimated0.2%
  5. #79gpt-oss-120b0.2%
No published score — estimated from related benchmarks
~#81 of 91 (estimated)~4.9%
  1. #1Muse Spark 1.350.5%
  2. #2GLM-5.350.3%
  3. #80Qwen3 Coder Next5.4%
  4. ~#81EXAONE 4.0 32BEstimated4.9%
  5. #81K2 Horizon 0.9B4.9%
No published score — estimated from related benchmarks