Understand this story
Cognition releases the fast coding model SWE-2 Flash, which is based on thinkymachines' Inkling-Small (276B total parameters, 12B activated parameters) and is 5x cheaper and 3x faster than SWE-2. Cognition says it surpasses all US open-weight models on major benchmarks, including models with 3x more parameters. On the training side, Cognition used large-scale reinforcement learning, making adjustments to the SWE-2 recipe; the main gains came from higher-quality and more diverse RL tasks. The team audited task instructions, environment configurations, and verifiers, and added several new task types. The model also adds a mid-training stage before RL for the first time, training on trillions of tokens of code, scientific articles, and blogs. Within this month, Pro, Max, and Teams users can use it for free in Devin Desktop and Devin CLI, and the weights will be opened through a secure sharing program with partners and researchers.
AI-generated from public reports
Latest development · Report
Cognition Releases SWE-2 Flash Coding ModelDevelopments
1 confirmed developments
- Source:Cognition
Reports and discussion timeline
Browse reports and discussion by time. Their evidence levels remain distinct.
Nov 5
CognitionFirst-partySelected
Cognition Releases SWE-2 Flash Coding Model
Cognition has released SWE-2 Flash, a fast and efficient coding model, claiming it outperforms all US open-weight models on major benchmarks, including models 3 times its size. The model is based on thinkymachines' Inkling-Small, with 276B total parameters and 12B activated parameters, and is 5 times cheaper and 3 times faster than SWE-2. Within this month, Pro, Max, and Teams users can use it for free in Devin Desktop and Devin CLI, and Cognition also plans to launch a partnership program for securely sharing weights.
CognitionFirst-party
Cognition Releases SWE-2 Flash Coding Model, Free in Devin This Month
Cognition has released SWE-2 Flash, a fast and efficient coding model, free this month for Pro, Max, and Teams users of Devin Desktop and Devin CLI. The model is based on thinkymachines' Inkling-Small, with 276B total parameters and 12B activated parameters; combined with an optimized inference stack, it costs 5 times less and runs 3 times faster than SWE-2. Cognition says it outperforms all US open-weight models on major benchmarks, and is building a partnership program for securely sharing weights.
CognitionFirst-partySelected
Cognition Releases SWE-2 Flash Coding Model and Discloses Training Details
Cognition releases the fast coding model SWE-2 Flash, based on thinkymachines' Inkling-Small (276B total parameters, 12B activated parameters), at 5x lower cost and 3x faster than SWE-2. Training uses large-scale RL, with adjustments made to the SWE-2 recipe, but the main gains come from higher-quality, more diverse RL tasks; the research team audited task instructions, environment configurations, and verifiers and added several new task types. The model also adds a mid-training stage before RL for the first time, trained on trillions of tokens of code, scientific articles, and blogs, and is available free of charge to Pro, Max, and Teams users in Devin Desktop and Devin CLI until the end of this month.
CognitionFirst-partySelected
Cognition Releases SWE-2 Flash Coding Model, Built on Inkling-Small with New…
Cognition has released the fast coding model SWE-2 Flash, built on thinkymachines' Inkling-Small (276B total parameters, 12B active parameters), at 5x lower cost and 3x faster than SWE-2. The model adds a mid-training stage for the first time before RL, training on trillions of tokens of code, scientific articles, and blogs, saving the scarcest high-quality data for last. Cognition says it surpasses all US open-weight models on major benchmarks, including models with 3x more parameters, and plans to safely share the weights with partners and researchers through the program; Pro, Max, and Teams users of Devin Desktop and Devin CLI can use it for free until the end of this month.
CognitionFirst-partySelected
Cognition Releases Coding Model SWE-2 Flash, Built on Inkling-Small
Cognition has released the fast coding model SWE-2 Flash, built on thinkymachines' Inkling-Small, with 276B total parameters and 12B activated parameters. The company claims it costs 5 times less and is 3 times faster than SWE-2, and surpasses all US open-weight models on major benchmarks, including models 3 times larger. The model will be free this month for Pro, Max, and Teams users of Devin Desktop and Devin CLI, and its weights will be released on a targeted basis through partner programs.
CognitionFirst-partySelected
Cognition Releases Fast Coding Model SWE-2 Flash
Cognition released the fast coding model SWE-2 Flash, claiming it surpasses all US open-weight models on major benchmarks, including models with 3 times the parameter count. The model is based on thinkymachines' Inkling-Small (276B total parameters, 12B active), paired with an optimized inference stack, and is 5 times cheaper and 3 times faster than SWE-2. Within this month, Pro, Max, and Teams users can use it for free in Devin Desktop and Devin CLI, and the weights will be opened through a secure sharing program with partners and researchers.