Skip to content
Model Pareto

Atria Dawn Preview

Shanghai AI Lab (Intern)Open weightshuggingface.co ↗
Input
—
Output
—
Speed
~81tok/sest.
Context
262Ktokens

Agentic model from Shanghai AI Lab built on GLM-5.2 (744B MoE). No API price found. Card also reports self-run Terminal-Bench 2.1, AutomationBench, GDPval and tau3-Banking, not mapped (our ids hold AA / Terminal-Bench 4.0 runs).

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
8× B200 180GB
FP8 weights
Throughput
~18,596 tok/s
many requests batched
Price
$0.30–2.69 /M tok
busy → light use
On your own machine
Mac Studio M5 Ultra 512GB (int4, ~42 tok/s single-stream)
single consumer GPU or Mac
Assumptions (5)
  • FP8 weights (744B params, 40B active per token) + 25% KV-cache headroom ≈ 930 GB
  • 8× B200 180GB at $5.98–$14.24/GPU-hour on-demand (2026-09-24)
  • ~18,596 output tok/s aggregate at batch 532 (bandwidth-bound); MoE compute scales with active params
  • Blended 3:1 input:output; prefill ~212,625 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

~#6 of 21 (estimated)~80.2%
  1. #1Claude Opus 596.0%
  2. #2Claude Sonnet 585.2%
  3. #5MiniMax-M380.5%
  4. ~#6Atria Dawn PreviewEstimated80.2%
  5. #6Hy378.0%
No published score — estimated from related benchmarks
SciCodeEstimated
~#27 of 119 (estimated)~55.5%
  1. #1Claude Fable 5.267.7%
  2. #2Claude Opus 5.566.9%
  3. #26GPT-6.1 Sol55.7%
  4. ~#27Atria Dawn PreviewEstimated55.5%
  5. #27Claude Haiku 5.555.0%
No published score — estimated from related benchmarks
~#9 of 26 (estimated)~53.4
  1. #1Claude Opus 5.561.7
  2. #2Claude Opus 559.7
  3. #8GPT-6.1 Sol54.1
  4. ~#9Atria Dawn PreviewEstimated53.4
  5. #9Gemini 4 Argon53.2
No published score — estimated from related benchmarks

Agents

~#3 of 15 (estimated)~83.3%
  1. #1Kimi K384.8%
  2. #2Qwen3.8 27B84.3%
  3. ~#3Atria Dawn PreviewEstimated83.3%
  4. #3Nex-N2.5-Pro82.2%
No published score — estimated from related benchmarks
OSWorld 2.0Estimated
~#3 of 5 (estimated)~57.9%
  1. #1GLM-5.3 Flash59.1%
  2. #2Kimi K358.3%
  3. ~#3Atria Dawn PreviewEstimated57.9%
  4. #3GPT-5.6 Terra50.2%
No published score — estimated from related benchmarks
~#22 of 92 (estimated)~1456 Elo
  1. #1Claude Opus 5.51866
  2. #2Muse Wonder Full1857
  3. #21GPT-6 Sol1487
  4. ~#22Atria Dawn PreviewEstimated1456
  5. #22Claude Sonnet 51449
No published score — estimated from related benchmarks
~#29 of 85 (estimated)~38.6%
  1. #1Claude Fable 5.276.4%
  2. #2Muse Wonder Full71.3%
  3. #28GPT-5.547.3%
  4. ~#29Atria Dawn PreviewEstimated38.6%
  5. #29K2 Horizon 375B A23B37.2%
No published score — estimated from related benchmarks
~#22 of 91 (estimated)~31.3%
  1. #1Muse Spark 1.350.5%
  2. #2GLM-5.350.3%
  3. #21K2 Horizon MoVA 36B A4B32.0%
  4. ~#22Atria Dawn PreviewEstimated31.3%
  5. #22Inkling29.1%
No published score — estimated from related benchmarks