Poolside

Laguna M.1

PoolsideBalanced

This model is no longer available on Elosia. This page is kept for informational purposes.

Performance Tier

Balanced

Laguna M.1 is a balanced model from Poolside : strong performance at a reasonable price.

Strong cost-performance ratio. Reliable for most professional use cases without premium pricing.

Capabilities

Context Length
Max Output Tokens

Benchmarks

General Intelligence
MMLU
Not reported
Mathematics
MATH-500
Not reported
Programming
HumanEval
Not reported
SWE-bench Verified
74.6%
SWE-bench Multilingual
63.1%
Agentic
SWE-bench Pro
49.2%
Terminal-Bench 2.0
45.8%

Where does Laguna M.1 stand?

Compare its performance index against every other model.

View the performance leaderboard

Recommended Use Cases

Coding

Strengths

  • Purpose-built for agentic coding (SWE-bench Verified 74.6%, SWE-bench Pro 49.2%)
  • Strong multilingual code performance (SWE-bench Multilingual 63.1%)
  • Efficient Mixture-of-Experts (225B total / 23B active) with a 256K-token context
  • Open weights under Apache 2.0

Limitations

  • Code specialist: no published general-knowledge or math benchmarks
  • Benchmarks are self-reported by Poolside (no independent leaderboard)

Frequently asked questions

Resources

This model may use your data for training

Similar Models