Skip to content
Model Pareto

Laguna S 2.1

poolsideOpen weightspoolside.ai ↗
Compare with nearest rivalsCompare with…
Input
—
Output
—
Speed
~99tok/sest.
Context
1Mtokens

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
2× B200 180GB
FP8 weights
Throughput
~15,523 tok/s
many requests batched
Price
$0.084–0.75 /M tok
busy → light use
On your own machine
Mac Studio M5 Max 128GB (int4, ~110 tok/s single-stream)
single consumer GPU or Mac
Assumptions (5)
  • FP8 weights (118B params, 8B active per token) + 40% KV-cache headroom ≈ 165 GB
  • 2× B200 180GB at $5.98–$14.24/GPU-hour on-demand (2026-09-24)
  • ~15,523 output tok/s aggregate at batch 444 (bandwidth-bound); MoE compute scales with active params
  • Blended 3:1 input:output; prefill ~265,781 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

SWE-bench VerifiedEstimated
~#3 of 14 (estimated)~81.1%
  1. #1Claude Opus 596.0%
  2. #2Claude Sonnet 585.2%
  3. ~#3Laguna S 2.1Estimated81.1%
  4. #3Gemini 3.1 Pro80.6%
No published score — estimated from related benchmarks
Terminal-Bench 4.0Estimated
~#30 of 89 (estimated)~11.1%
  1. #1Claude Sonnet 5.563.6%
  2. #2Claude Opus 5.559.6%
  3. #27Kimi K312.6%
  4. ~#30Laguna S 2.1Estimated11.1%
  5. #30Qwen3.8 2.4T A95B11.1%
No published score — estimated from related benchmarks
LiveCodeBenchEstimated
~#10 of 25 (estimated)~77.4%
  1. #1Qwen3.8-Omni-Flash92.6%
  2. #2Qwen3.8 Flash-Next91.9%
  3. #9gpt-oss-20b77.7%
  4. ~#10Laguna S 2.1Estimated77.4%
  5. #10Bonsai 27B76.4%
No published score — estimated from related benchmarks
SciCodeEstimated
~#37 of 106 (estimated)~49.4%
  1. #1Claude Opus 5.566.9%
  2. #2Claude Fable 5.163.1%
  3. #36Inkling Small49.7%
  4. ~#37Laguna S 2.1Estimated49.4%
  5. #37Hy348.6%
No published score — estimated from related benchmarks
AA Coding Agent IndexEstimated
~last of 18 (estimated)below measured range (<41.1)
  1. #1Claude Opus 5.566.0
  2. #2Gemini 4 Argon63.8
  3. #17GPT-6 Luna41.1
  4. ~#18Laguna S 2.1Estimated<41.1
No published score — estimated from related benchmarks