Cyborg Sonic 3.0

Cyborg Sonic 3.0

Cyborg Sonic 3.0

Our hybrid reasoning, fastest and intelligence model yet, with built-in thinking that puts expert-level intelligence in everyone’s hands.

Our hybrid reasoning, fastest and intelligence model yet, with built-in thinking that puts expert-level intelligence in everyone’s hands.

introduction

introduction

We are introducing Sonic-3.0, our best AI system yet. Sonic-3.0 is a significant leap in intelligence over all our previous models, featuring state-of-the-art performance across coding, math, writing, health, visual perception, and more. It is a unified system that knows when to respond quickly and when to think longer to provide expert-level responses. Sonic-3.0 is available to all users, with Pro subscribers getting access to Sonic-3.0 pro, a version with extended reasoning for even more comprehensive and accurate answers.

We are introducing Sonic-3.0, our best AI system yet. Sonic-3.0 is a significant leap in intelligence over all our previous models, featuring state-of-the-art performance across coding, math, writing, health, visual perception, and more. It is a unified system that knows when to respond quickly and when to think longer to provide expert-level responses. Sonic-3.0 is available to all users, with Pro subscribers getting access to Sonic-3.0 pro, a version with extended reasoning for even more comprehensive and accurate answers.

Cyborg Sonic 3.0 July, 2025

Cyborg Sonic 3.0 July, 2025

Sonic-3.0 is a drop-in replacement for Sonic 1.5 that delivers superior performance and precision for real-world coding and agentic tasks. It handles complex, multi-step problems with more rigor and attention to detail.

Sonic-3.0 is a drop-in replacement for Sonic 1.5 that delivers superior performance and precision for real-world coding and agentic tasks. It handles complex, multi-step problems with more rigor and attention to detail.

Unified system

Unified system

Sonic-3.0 is a unified system with a smart, efficient model that answers most questions, a deeper reasoning model (Sonic-3.0 thinking) for harder problems, and a real‑time router that quickly decides which to use based on conversation type, complexity, tool needs, and your explicit intent (for example, if you say “think hard about this” in the prompt). The router is continuously trained on real signals, preference rates for responses, and measured correctness, improving over time. Once usage limits are reached, a mini version of each model handles remaining queries. In the near future, we plan to integrate these capabilities into a single model.

Sonic-3.0 is a unified system with a smart, efficient model that answers most questions, a deeper reasoning model (Sonic-3.0 thinking) for harder problems, and a real‑time router that quickly decides which to use based on conversation type, complexity, tool needs, and your explicit intent (for example, if you say “think hard about this” in the prompt). The router is continuously trained on real signals, preference rates for responses, and measured correctness, improving over time. Once usage limits are reached, a mini version of each model handles remaining queries. In the near future, we plan to integrate these capabilities into a single model.

Coding

Coding

Sonic-3.0 is our strongest coding model to date. It shows particular improvements in complex front‑end generation and debugging larger repositories. It can often create beautiful and responsive websites, apps, and games with an eye for aesthetic sensibility in just one prompt, intuitively and tastefully turning ideas into reality. Early testers also noted its design choices, with a much better understanding of things like spacing, typography, and white space.

Sonic-3.0 is our strongest coding model to date. It shows particular improvements in complex front‑end generation and debugging larger repositories. It can often create beautiful and responsive websites, apps, and games with an eye for aesthetic sensibility in just one prompt, intuitively and tastefully turning ideas into reality. Early testers also noted its design choices, with a much better understanding of things like spacing, typography, and white space.

Here are a examples of what Sonic-3.0 has created with just one prompt:

Here are a examples of what Sonic-3.0 has created with just one prompt:

Creative expression and writing

Creative expression and writing

Sonic-3.0 is our most capable writing collaborator yet, able to help you steer and translate rough ideas into compelling, resonant writing with literary depth and rhythm. It more reliably handles writing that involves structural ambiguity, such as sustaining unrhymed iambic pentameter or free verse that flows naturally, combining respect for form with expressive clarity. These improved writing capabilities mean that Cyborg Ai is better at helping you with everyday tasks like drafting and editing reports, emails, memos, and more.

Sonic-3.0 is our most capable writing collaborator yet, able to help you steer and translate rough ideas into compelling, resonant writing with literary depth and rhythm. It more reliably handles writing that involves structural ambiguity, such as sustaining unrhymed iambic pentameter or free verse that flows naturally, combining respect for form with expressive clarity. These improved writing capabilities mean that Cyborg Ai is better at helping you with everyday tasks like drafting and editing reports, emails, memos, and more.

Evaluations

Evaluations

Sonic-3.0 is much smarter across the board, as reflected by its performance on academic and human-evaluated benchmarks, particularly in math, coding, visual perception, and health. It sets a new state of the art across math (86.6% on AIME 2025 without tools), real-world coding (68.9% on SWE-bench Verified, 78% on Aider Polyglot), multimodal understanding (84.2% on MMMU), and health (46.2% on HealthBench Hard)—and those gains show up in everyday use. With Sonic-3.0 pro’s extended reasoning, the model also sets a new SOTA on GPQA, scoring 88.4% without tools.

Sonic-3.0 is much smarter across the board, as reflected by its performance on academic and human-evaluated benchmarks, particularly in math, coding, visual perception, and health. It sets a new state of the art across math (86.6% on AIME 2025 without tools), real-world coding (68.9% on SWE-bench Verified, 78% on Aider Polyglot), multimodal understanding (84.2% on MMMU), and health (46.2% on HealthBench Hard)—and those gains show up in everyday use. With Sonic-3.0 pro’s extended reasoning, the model also sets a new SOTA on GPQA, scoring 88.4% without tools.

AIME 2025 Competition math

AIME 2025 Competition math

*AIME results with tools should not be compared directly to the performance of models without tool access; they are an example of how effectively Sonic-3.0 leverages available tools.

*AIME results with tools should not be compared directly to the performance of models without tool access; they are an example of how effectively Sonic-3.0 leverages available tools.

Coding SWE Bench

Coding SWE Bench

All SWE-bench evaluation runs use a fixed subset of n=477 verified tasks which have been validated on our internal infrastructure.

All SWE-bench evaluation runs use a fixed subset of n=477 verified tasks which have been validated on our internal infrastructure.

GPQA DiamondPhD-level science questions

GPQA DiamondPhD-level science questions