When was SWE-2 Flash released?

SWE-2 Flash was released on November 4, 2026 by cognition. This is the official SWE-2 Flash release date tracked on LLM Stats.

How big is SWE-2 Flash?

SWE-2 Flash has 276 billion parameters.

Who created SWE-2 Flash?

SWE-2 Flash was created by cognition.

What is the license for SWE-2 Flash?

SWE-2 Flash is released under the Proprietary license.

Where is the SWE-2 Flash paper or technical report?

SWE-2 Flash has a paper or technical report available at https://cognition.com/blog/swe-2-flash. Use that source for architecture, training, release and evaluation details.

What models should I compare SWE-2 Flash against?

Common SWE-2 Flash comparisons include SWE-2 Flash vs Muse Spark 1.3, SWE-2 Flash vs Claude Opus 4.8, SWE-2 Flash vs GPT-5.5. Compare them side by side for benchmark scores, pricing, context window, latency and provider availability.

SWE-2 Flash Benchmarks, Pricing & Context Window

Name: SWE-2 Flash
Author: cognition

SWE-2 Flash: Benchmarks, Pricing & Context Window

SWE-2 Flash is a language model from Cognition, released in November 2026.

SWE-2 Flash is Cognition's fast, low-cost coding model, built on Thinking Machines' Inkling-Small (276B total / 12B active parameters) with a large-scale mid-training stage (a few trillion tokens of scientific articles, blogs, code and

SWE-2 Flash benchmarks

Capability tiers

Standing within each category, adjusted for leaderboard depth.

Real tasks performance

High-confidence performance for SWE-2 Flash across real-world prompt categories. Only 95% intervals at most 4 points wide are shown.

Performance by conversation depth

How SWE-2 Flash holds up as conversations get longer.

Quality Tracker

SWE-2 Flash Performance Across Datasets

Scores sourced from the model's scorecard, paper, or official blog posts

llm-stats.com - Thu Nov 12 2026

Notice missing or incorrect data?

Try now

huggle

SWE-2 Flashin Huggle

Make it with
SWE-2 Flash.

SWE-2 Flash latency

SWE-2 Flash time to first token, sustained output throughput, and failed-request rate from live model usage over the trailing 7 days.

SWE-2 Flash examples

Recent arena outputs from SWE-2 Flash, picked from the highest-ranked matchups.

SWE-2 Flash license

SWE-2 Flash is a proprietary model available under its provider's product and API terms, has 276.0B parameters.

License: Proprietary; Hosted access
Parameters: 276.0B

Proprietary license - usage restrictions apply

SWE-2 Flash resources

Official sources for SWE-2 Flash: official playground, official launch post.

SWE-2 Flash vs other models

The most-compared alternatives to SWE-2 Flash are Muse Spark 1.3, Claude Opus 4.8, GPT-5.5. Open any pair side-by-side for benchmarks, pricing, context, and latency.