Skip to main content

Model profile

Laguna S 2.1

PoolsideCurrentReleased Jul 21, 2026
Data verified
Overall Score
Unranked
Arena Elo
Not listed
Eligible category ranks
0of 8
Price (1M tokens)
$0.1 in / $0.2 out
API pricing
Speed
Not listed
Context
1M

Evidence coverage

6 of 323 tracked benchmarks are published. 6 are verified and 0 provisional. 2 of 8 categories are measured.

Updated July 21, 2026Methodology
Published / tracked
6 / 323
Verified
6
Provisional
0
Categories with evidence
2 / 8

Evidence by category

  • Agentic2 benchmarks
    Verified
  • Coding4 benchmarks
    Verified
  • Reasoning0 benchmarks
    Not measured
  • Knowledge0 benchmarks
    Not measured
  • Math0 benchmarks
    Not measured
  • Multilingual0 benchmarks
    Not measured
  • Multimodal0 benchmarks
    Not measured
  • Inst. Following0 benchmarks
    Not measured
Open WeightSelf-hostReasoning
Confidence:
Low
s-2-1

BenchLM is tracking Laguna S 2.1, but this profile is currently excluded from the public leaderboard because it still lacks enough non-generated benchmark coverage to rank safely. Only non-generated public benchmark rows appear below.

Laguna S 2.1 is a open weight model with a 1M token context window. It uses explicit chain-of-thought reasoning, which typically improves performance on math and complex reasoning tasks at the cost of higher latency and token usage.

Open weights under OpenMDW-1.1 are available on Hugging Face. Poolside also lists hosted access through OpenRouter, Baseten, Kilo, and its pool coding agent.

Laguna S 2.1 sits inside the Laguna family alongside Laguna M.1, Laguna XS.2. BenchLM links it directly to Laguna M.1 as the earlier related model in that lineage. This profile currently has 6 of 323 tracked benchmarks. BenchLM only exposes non-generated benchmark rows publicly, so missing categories stay blank until a sourced evaluation is available.

Peer position

Exact provisional scores and ranks for the closest listed peers. A score can appear before a model clears the evidence threshold for a rank, so equal scores can have different rank states.

Range 77.4483.93

  1. Claude Mythos 5
    Anthropic
    #183.93
    Claude Mythos 5 is #1 with a score of 83.93.
    Compare
  2. Claude Fable 5
    Anthropic
    #283.68
    Claude Fable 5 is #2 with a score of 83.68.
    Compare
  3. GPT-5.6 Sol
    OpenAI
    #381.96
    GPT-5.6 Sol is #3 with a score of 81.96.
    Compare
  4. Kimi K3
    Moonshot AI
    #480.96
    Kimi K3 is #4 with a score of 80.96.
    Compare
  5. Claude Opus 4.8
    Anthropic
    #578.34
    Claude Opus 4.8 is #5 with a score of 78.34.
    Compare
  6. Muse Spark 1.1
    Meta
    #677.44
    Muse Spark 1.1 is #6 with a score of 77.44.
    Compare
  7. Laguna S 2.1Current model
    Poolside
    UnrankedNot measured
    Laguna S 2.1 is Unranked with a score of Not measured.

Category percentile

More

Relative position among models eligible for each sourced category. A higher percentile means a stronger position within that category's ranked cohort; 100 is highest.

No eligible category percentile is available from the published evidence yet.

Category evidence

Scores and ranks appear only where this model has published benchmark evidence. Categories without displayable source records remain not measured.

Category scores, ranks, weighting, benchmark coverage, and evidence status
CategoryScore
AgenticRank Not rankedWeight 22%2 benchmarksVerified64.8
CodingRank Not rankedWeight 20%4 benchmarksVerified48.5
ReasoningWeight 17%0 benchmarksNot measuredNot measured
KnowledgeWeight 12%0 benchmarksNot measuredNot measured
MathWeight 5%0 benchmarksNot measuredNot measured
MultilingualWeight 7%0 benchmarksNot measuredNot measured
MultimodalWeight 12%0 benchmarksNot measuredNot measured
Inst. FollowingWeight 5%0 benchmarksNot measuredNot measured

Benchmark Details

Rows below have a displayable published verification record. Each source link and provenance note remains in the page HTML while its category is closed. Source-unverified manual rows and generated rows stay hidden.

Agentic2 benchmarks
Terminal-Bench 2.0Provider exact
70.2%Weighted 38%
Source: Poolside: Introducing Laguna S 2.1Provenance: Poolside reports 70.2% on Terminal-Bench 2.1 with thinking enabled in the pool harness, averaged over four attempts per task. BenchLM stores the value on its existing Terminal-Bench key.
Toolathlon-VerifiedProvider exact
49.7%Display only
Source: Poolside: Introducing Laguna S 2.1Provenance: Poolside reports 49.7% on Toolathlon Verified with a custom Toolathlon agent, averaged over three runs.
Coding4 benchmarks
SWE-bench ProProvider exact
59.4%Weighted 10%
Source: Poolside: Introducing Laguna S 2.1Provenance: Poolside reports 59.4% on the SWE-bench Pro public dataset with the Harbor adapter and pool harness, averaged over four attempts per task.
Terminal-Bench 2.0Provider exact
70.2%Display only
Source: Poolside: Introducing Laguna S 2.1Provenance: Poolside reports 70.2% on Terminal-Bench 2.1 with thinking enabled in the pool harness, averaged over four attempts per task. BenchLM stores the value on its existing Terminal-Bench key.
SWE MultilingualProvider exact
78.5%Display only
Source: Poolside: Introducing Laguna S 2.1Provenance: Poolside reports 78.5% on SWE-bench Multilingual with the Harbor adapter and pool harness, averaged over four attempts per task.
deepSweProvider exact
40.4%Display only
Source: Poolside: Introducing Laguna S 2.1Provenance: Poolside reports 40.4% on DeepSWE v1.1 with thinking enabled in the pool harness, averaged over three attempts per task. BenchLM keeps DeepSWE display-only because the value combines the model with an agent harness.

Laguna Family

s-2-1 · 118B-A8B

Related Earlier Model

Laguna M.1

Frequently Asked Questions

How does Laguna S 2.1 perform overall in AI benchmarks?

Laguna S 2.1 has 6 published benchmark scores on BenchLM, but it does not yet have enough non-generated coverage to receive a global overall rank.

Is Laguna S 2.1 good for coding and programming?

Laguna S 2.1 has visible benchmark coverage in coding and programming, but BenchLM does not currently assign it a global category rank there.

Is Laguna S 2.1 good for agentic tool use and computer tasks?

Laguna S 2.1 has visible benchmark coverage in agentic tool use and computer tasks, but BenchLM does not currently assign it a global category rank there.

Is Laguna S 2.1 open source?

Yes, Laguna S 2.1 is an open weight model created by Poolside, meaning it can be downloaded and run locally or fine-tuned for specific use cases.

Which sibling models are related to Laguna S 2.1?

Laguna S 2.1 belongs to the Laguna family. Related variants on BenchLM include Laguna M.1, Laguna XS.2.

Does Laguna S 2.1 have full benchmark coverage on BenchLM?

Not yet. Laguna S 2.1 currently has 6 published benchmark scores out of the 323 benchmarks BenchLM tracks. BenchLM only exposes non-generated public benchmark rows, so missing categories stay blank until a sourced evaluation is available.

What is the context window size of Laguna S 2.1?

Laguna S 2.1 has a published context window of 1M, which determines how much text it can process in a single interaction.

Last updated: July 21, 2026 · Runtime metrics stay blank until BenchLM has a sourced snapshot.

Choose with this week’s evidence

Join 2,000+ readers for ranking moves, new releases, pricing changes, and the evidence behind them.

Free. One email per week.