llmboard.aiAI model intelligence
Home

Model Rankings

OverallOpen ModelsAgentCodingReasoningMathKnowledgeInstruction FollowingTextVision
Image GenerationImage Editing
Video GenerationImage to VideoVideo Editing
Text to SpeechSpeech to Text
Embeddings

Efficiency

Chat Token PricingImage PricingVideo PricingAudio Pricing
Chat Speed & LatencyProvider Reliability

Benchmarks

GPQAMMLU-ProAIME 2025SWE-Bench VerifiedMMLUHumanity's Last ExamLiveCodeBenchMATHHumanEvalMMMU-Pro
All Benchmarks

Tools

Model Directory

Scoring & Data

Scoring & Data
1464 models816 benchmarks

Leaderboard Center

Overall RankingCodingCore BenchmarksPrice & ValueRuntime Performance

Modalities

All ModelsImage GenerationImage EditingVideo GenerationImage-to-VideoVideo EditingText-to-SpeechSpeech-to-TextEmbeddings

Data & Methods

Scoring MethodAll Benchmarks
llmboard.aiCopyright 2026 llmboard.ai

Cognition model product

SWE 2 Flash

SWE 2 Flash is Cognition's coding model, built on Thinking Machines' Inkling-Small.

Updated Nov 22, 2026. Default version: SWE-2 Flash

LLMBoard Score69.9SWE-2 Flash
Coverage20%7 benchmark families
Context windowN/ATokens
Official input priceN/AOfficial price unavailable

On this page

  • Capability
  • Benchmarks
  • Arena
  • Pricing
  • Runtime
  • Specification
  • Versions
  • Similar models
  • About
  • FAQ

SWE 2 Flash Capability Profile

This profile uses the model's current scored version. Arena ratings and prices are shown separately.

SWE-2 Flash LLMBoard score breakdown

SWE 2 Flash Benchmark Results

Benchmark scores for SWE-2 Flash.

7 rows
Columns

Show columns

Sort by
Benchmark
Score
Rank
Participants
Percentile
Evidence
Evaluated
BenchmarkTerminal-Bench 2.1Score89.20%Rank05Participants58Percentile92.98%EvidenceCEvaluatedNov 21, 2026
BenchmarkFrontierCode 2.0Score24.00%Rank05Participants7Percentile33.33%EvidenceCEvaluatedNov 21, 2026
BenchmarkSWE Atlas - Codebase QnAScore50.00%Rank06Participants10Percentile44.44%EvidenceCEvaluatedNov 21, 2026
BenchmarkExploitBenchScore34.80%Rank10Participants12Percentile18.18%EvidenceCEvaluatedNov 21, 2026
BenchmarkFrontierCode 1.1Score45.50%Rank12Participants25Percentile54.17%EvidenceCEvaluatedNov 21, 2026
BenchmarkTerminal-Bench 4.0Score20.00%Rank34Participants41Percentile17.50%EvidenceCEvaluatedNov 21, 2026
BenchmarkDeepSWE 1.1Score60.10%Rank39Participants59Percentile34.48%EvidenceCEvaluatedNov 21, 2026

SWE 2 Flash Arena Results

Preference and agent-evaluation results for the default version.

No Arena results

The default version does not have a matching Arena result yet.

SWE 2 Flash Pricing

Official vendor API pricing appears first, followed by individual provider offers.

Official API
N/A
Official provider
N/A
Lowest third-party
N/A
Tracked offerings
0
No provider prices

The default version has no current input or output token prices.

Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.

SWE 2 Flash Runtime Performance

Provider-specific output speed and catalog latency for SWE-2 Flash. Runtime does not affect the capability score.

No runtime data

No provider-specific speed or latency record is linked to the default version yet.

Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.

SWE 2 Flash Specifications

Technical details for the model's default version.

Version
SWE-2 Flash
Released
Nov 4, 2026
Knowledge cutoff
Unknown
Parameters
276B
Context window
N/A
Max output
N/A
Inputs
text
Outputs
text
Open weights
No
License
Proprietary

SWE 2 Flash Versions

Available versions of this model. The score column identifies the version used in the overall ranking.

1 row
Columns

Show columns

Sort by
Version
Released
LLMBoard
Parameters
Context
Max output
Open weights
License
VersionSWE-2 FlashReleasedNov 4, 2026LLMBoard69.92Parameters276BContextN/AMax outputN/AOpen weightsNoLicenseProprietary

Models similar to SWE 2 Flash

Recommendations prioritize the same model type and family, then the closest LLMBoard score.

#44+5.71
CO

SWE 2

Cognition

75.63 LLMBoard

Details
#72-0.03
ME

Muse Spark

Meta

69.89 LLMBoard

Details
#73-0.09
IN

Ling 3.0 Flash Fin

Inclusionai

69.83 LLMBoard

Details
#74-0.27
AN

Claude Mythos 5.1

Anthropic

69.65 LLMBoard

Details
#70+0.33
RA

Beam 1.0

Reflection AI

70.25 LLMBoard

Details
#75-0.36
OP

GPT-5.2-Pro

OpenAI

69.56 LLMBoard

Details

What is SWE 2 Flash?

Key information about SWE 2 Flash and its available data.

It uses Inkling-Small (276B total / 12B active parameters), with a mid-training stage followed by reinforcement learning. Starting 2026-11-04, it is available in Devin Desktop and Devin CLI; its weights are not public.

Data as of 2026-11-21.

FAQ

Common questions about SWE 2 Flash.

When was SWE 2 Flash released?

SWE 2 Flash's default version was released on Nov 4, 2026.

How much does SWE 2 Flash cost?

No official standard PAYG price is currently available for SWE 2 Flash.

Who created SWE 2 Flash?

SWE 2 Flash was created by Cognition.

What is the context window for SWE 2 Flash?

A context window is not available for the default version.

Is SWE 2 Flash open weight?

No. The default version is not marked as having publicly available weights.

How many API providers offer SWE 2 Flash?

No provider offering is currently linked to the default version.

Browse runtime rankings