EXAONE 4.5 33B
- Input
- —
- Output
- —
- Speed
- ~83tok/sest.
- Context
- 262Ktokens
Long-tail import from Artificial Analysis (2026-09-24); AA variant: EXAONE 4.5 33B. Price = AA median host price.
Rankings
Writing & chat
Verified#73 / 162rank by quality49.6±9.5quality / 1001/5 benchmarks measuredVision
Verified#63 / 98rank by quality46.3±8.2quality / 1001/6 benchmarks measuredHard reasoning
Verified#109 / 218rank by quality43.3±5.9quality / 1002/4 benchmarks measuredAgents
Verified#69 / 187rank by quality40.6±7.8quality / 1002/7 benchmarks measured
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
- ~4,788 tok/s
- many requests batched
- Price
- $0.17–1.49 /M tok
- busy → light use
- On your own machine
- RTX 5090 32GB (int4, ~76 tok/s single-stream)
- single consumer GPU or Mac
Assumptions (5)
- FP8 weights (33B params) + 25% KV-cache headroom ≈ 41 GB
- 1× B200 180GB at $5.98–$14.24/GPU-hour on-demand (2026-09-24)
- ~4,788 output tok/s aggregate at batch 137 (bandwidth-bound)
- Blended 3:1 input:output; prefill ~32,216 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
IFBenchVerified
#61 of 13958.0%
- #1MiniMax-M382.9%
- #2Qwen3.8 2.4T A95B82.8%
- #60NVIDIA Nemotron 3 Nano 4B58.2%
- #61EXAONE 4.5 33B58.0%
- #62Solar Open 100B57.7%
artificialanalysis.ai · 2026-09-24
LMArena Text, Non-English (Elo)Estimated
~#42 of 57 (estimated)~1394 Elo
- #1Gemini 4 ArgonFew votes1512
- #2Claude Opus 5.51497
- #41Claude Haiku 4.51395
- ~#42EXAONE 4.5 33BEstimated1394
- #42GPT-5.4 nano1387
No published score — estimated from related benchmarks
LMArena Text (Elo)Estimated
~#42 of 57 (estimated)~1411 Elo
- #1Gemini 4 Argon1525
- #2Claude Opus 5.51507
- #41Mistral Large 31413
- ~#42EXAONE 4.5 33BEstimated1411
- #42GPT-5.4 nano1402
No published score — estimated from related benchmarks
EQ-Bench Creative Writing v3 (Elo)Estimated
~#16 of 26 (estimated)~1725 Elo
- #1GPT-6 Astra2173
- #2Claude Fable 5.12162
- #15Gemini 3.8 Flash1748
- ~#16EXAONE 4.5 33BEstimated1725
- #16Nemotron 3 Ultra1692
No published score — estimated from related benchmarks
Vision
MMMU-ProVerified
#51 of 8767.3%
- #1Claude Opus 5.587.7%
- #2GPT-6 Astra86.9%
- #50Ternary Bonsai 27B69.0%
- #51EXAONE 4.5 33B67.3%
- #52DiffusionGemma 26B A4B66.5%
artificialanalysis.ai · 2026-09-24
CharXiv ReasoningEstimated
~#11 of 12 (estimated)~75.2%
- #1Gemini 3.8 Flash86.2%
- #2Kimi K384.8%
- #10Claude Sonnet 577.0%
- ~#11EXAONE 4.5 33BEstimated75.2%
- #11Gemini 3.5 Flash-Lite74.5%
No published score — estimated from related benchmarks
OmniDocBench v1.5Estimated
~#5 of 8 (estimated)~81.8%
- #1MiniMax-M391.6%
- #2Kimi K391.1%
- #4Gemini 3.1 Pro85.3%
- ~#5EXAONE 4.5 33BEstimated81.8%
- #5Claude Haiku 4.579.6%
No published score — estimated from related benchmarks
GDP.pdf (AA)Estimated
~#34 of 48 (estimated)~9.8%
- #1Muse Wonder Full32.6%
- #2GPT-6 Astra31.0%
- #33Muse Glimmer10.0%
- ~#34EXAONE 4.5 33BEstimated9.8%
- #34MiniMax-M39.8%
No published score — estimated from related benchmarks
ChartographyEstimated
~#22 of 25 (estimated)~10.9%
- #1Claude Fable 5.273.4%
- #2GPT-6 Astra71.0%
- #21DeepSeek V4.1 Flash11.6%
- ~#22EXAONE 4.5 33BEstimated10.9%
- #22Gemini 3.5 Flash-Lite10.2%
No published score — estimated from related benchmarks
Hard reasoning
GPQA DiamondVerified
#80 of 19279.4%
- #1GPT-6 Astra96.1%
- #2Gemini 3.8 Flash95.3%
- #79Nemotron 3 Super 120B A12B80.0%
- #80EXAONE 4.5 33B79.4%
- #81Gemma 4 26B A4B79.2%
artificialanalysis.ai · 2026-09-24
Humanity's Last ExamVerified
#97 of 19812.9%
- #1Claude Fable 5.274.3%
- #2Claude Opus 5.561.4%
- #96INTELLECT-313.1%
- #97EXAONE 4.5 33B12.9%
- #98Cogito v2.112.0%
artificialanalysis.ai · 2026-09-24
AIME (latest)Estimated
~#23 of 34 (estimated)~77.1%
- #1GLM-5.3 NVFP499.5%
- #2Beam 1.0 501B A23B97.7%
- #22Phi-4-reasoning-plus78.0%
- ~#23EXAONE 4.5 33BEstimated77.1%
- #23Mistral Large 338.0%
No published score — estimated from related benchmarks
AA Intelligence IndexEstimated
~#65 of 103 (estimated)~18.7
- #1Gemini 4 Carbon59.4
- #2Claude Opus 5.557.6
- #64LongCat 2.019.1
- ~#65EXAONE 4.5 33BEstimated18.7
- #65Qwen3.5 397B A17B18.4
No published score — estimated from related benchmarks
Agents
τ²-bench (Telecom)Verified
#37 of 11978.1%
- #1Step 3.7 Flash98.5%
- #2DeepSeek-V4-Pro-0813-nvfp4-DSpark98.3%
- #35Qwen3 Coder Next79.5%
- #37EXAONE 4.5 33B78.1%
- #38GPT-5.4 nano76.0%
artificialanalysis.ai · 2026-09-24
OSWorld-VerifiedEstimated
~#11 of 15 (estimated)~66.4%
- #1Kimi K384.8%
- #2Qwen3.8 27B84.3%
- #10Nex-N2.5-Mini71.2%
- ~#11EXAONE 4.5 33BEstimated66.4%
- #11Muse Glimmer65.9%
No published score — estimated from related benchmarks
OSWorld 2.0Estimated
~last of 5 (estimated)below measured range (<47.5%)
No published score — estimated from related benchmarks
BrowseCompEstimated
~#19 of 22 (estimated)~66.8%
- #1Atria Dawn Preview92.5%
- #2GPT-6 Astra91.5%
- #18Nemotron 3.5 Bolt 74B-A9B70.0%
- ~#19EXAONE 4.5 33BEstimated66.8%
- #19Mistral Medium 3.548.6%
No published score — estimated from related benchmarks
GDPval-AA (Elo)Verified
#74 of 91458 Elo
- #1Claude Opus 5.51866
- #2Muse Wonder Full1857
- #73Mercury 2.5462
- #74EXAONE 4.5 33B458
- #75Nova 2.0 Pro Preview456
artificialanalysis.ai · 2026-09-24
AutomationBench-AAEstimated
~#45 of 85 (estimated)~17.1%
- #1Claude Fable 5.276.4%
- #2Muse Wonder Full71.3%
- #44Qwen3.7 Plus17.4%
- ~#45EXAONE 4.5 33BEstimated17.1%
- #45Ling-3.0-flash-VL15.7%
No published score — estimated from related benchmarks
τ-Bench Banking (AA)Estimated
~#46 of 91 (estimated)~16.8%
- #1Muse Spark 1.350.5%
- #2GLM-5.350.3%
- #45Nemotron 3.5 Bolt 74B-A9B17.0%
- ~#46EXAONE 4.5 33BEstimated16.8%
- #46A.X-K216.1%
No published score — estimated from related benchmarks