Skip to content
Model Pareto

Cogito v2.1

Deep CogitoOpen weightsartificialanalysis.ai ↗
Input
$1.25/M tok
Output
$1.25/M tok
Blended
$1.25/M tok · 3:1
Speed
~81tok/sest.
Context
128Ktokens

Long-tail import from Artificial Analysis (2026-09-24); AA variant: Cogito v2.1 (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, compared with the API price above.

Hardware
8× B200 180GB
FP8 weights
Throughput
~21,907 tok/s
many requests batched
Price
$0.26–2.32 /M tok
busy → light use
On your own machine
Mac Studio M5 Ultra 512GB (int4, ~45 tok/s single-stream)
single consumer GPU or Mac
Assumptions (5)
  • FP8 weights (671B params, 37B active per token) + 25% KV-cache headroom ≈ 839 GB
  • 8× B200 180GB at $5.98–$14.24/GPU-hour on-demand (2026-09-24)
  • ~21,907 output tok/s aggregate at batch 627 (bandwidth-bound); MoE compute scales with active params
  • Blended 3:1 input:output; prefill ~229,865 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