Observed Signal · Nov 4, 2026 · Product Launch · Source: Cognition · Impact: 2/5
Introducing SWE-2 Flash
Cognition has introduced SWE-2 Flash, an American coding model optimized for low cost and high speed. Built upon Thinking Machines' Inkling-Small base model, it incorporates a mid-training stage prior to reinforcement learning, substantially raising performance on Terminal-Bench 4.0 from 0.9% to 20.0% and FrontierCode 2.0 from 3.2% to 24.0%. Cognition also announced plans to securely share the model weights with select researchers and industry partners.
The release demonstrates significant coding efficiency gains by combining mid-training with reinforcement learning on open base models, challenging larger proprietary architectures in software engineering automation.
Track Cognition Signals & Market Shifts in Real-Time
Polaris7 autonomous intelligence agents track regulatory filings, primary sources, executive changes, and deal flow 24/7. Create your Polaris7 workspace to monitor these entities.
Key Takeaways & Evidence Grounding
- SWE-2 Flash introduces a mid-training stage before reinforcement learning into Cognition's training pipeline.
- Built on Thinking Machines' Inkling-Small base model, SWE-2 Flash improves Terminal-Bench 4.0 performance from 0.9% to 20.0%.
- On FrontierCode 2.0, the model reaches 24.0% compared to 3.2% for the Inkling-Small baseline.
- Cognition reports that SWE-2 Flash outperforms all American open-weight models, including those up to three times larger.
- Model weights will be distributed to industry partners and researchers through a secure sharing program.
Connected Companies & Entities
1 Entity mapped“Cognition developed and announced the SWE-2 Flash coding model.”
Related Market Signals & Shifts
Recent verified developments and strategic activity across this market segment.
MiniMax M3.1 Flash: Redefining Agent Performance and Cost
Today, we officially roll out MiniMax M3.1 Flash. As a next-generation Flash model, M3.1 Flash delivers a generational leap in agentic tasks, coding, Computer Use and cybersecurity, while making online inference nearly three times faster and reducing usage costs by 70%.
DeepSeek V4.1 Flash Beats Larger Model, Proving Ilya Right
Ilya Sutskever left OpenAI to found SSI and predicted the end of scaling. DeepSeek's new V4.1 Flash model, with a third of the parameters of its flagship V4-Pro, outperforms it on coding benchmarks (74.2 vs 62.7) while activating fewer parameters per token. Starting September 14, all V4-Pro requests route to V4.1 Flash. This validates the shift from model size to efficiency, marking a turning point in AI development.
DeepSeek Launches V4.1 Flash with Novel Encoder-Decoder Architecture
DeepSeek released DeepSeek-V4.1-Flash, a 763B-parameter mixture-of-experts model employing a novel causal encoder-decoder architecture with 8B active parameters for prefill and 16B for decode. It features native vision understanding, 1M token context, an MIT license, and extreme inference efficiency, claiming up to 1/8 KV cache footprint versus V4 Flash. Independent evals (Artificial Analysis Index 40, Vals Index #1 open-weight) show it surpasses V4 Pro at lower cost. API pricing is $0.30/1M input and $1.20/1M output tokens. DeepSeek has soft-retired V4 Pro, routing traffic to V4.1 Flash. The model supports SSD offload and local deployment, with Ollama and Baseten offering day-0 support. Technical discussions highlight the architecture's novelty and potential impact on long-context agents.
Receive all future signals for Cognition & competitors in your briefing
Never miss material M&A deals, leadership changes, account wins, or product rollouts. Polaris7 monitors 20,000+ primary sources 24/7, delivering verified telemetry directly into your executive briefing.