# Addendum to GPT-6 Astra System Card: GPT-6.1 Sol

Date: 29 Sep 2026
Released by: OpenAI

## MediaRelease.co Summary

OpenAI published a system card addendum for GPT-6.1 Sol covering updated safety metrics and preparedness ratings. It treats the model as Critical for cybersecurity and High for biological and chemical capability, using Astra’s safeguards stack.

## Original release

1. Introduction Today, we’re introducing GPT-6.1 Sol. GPT-6.1 is the latest model family in the GPT-6 series. GPT-6.1 Sol delivers capabilities comparable to those of our most powerful model, GPT-6 Astra, with an unmatched combination of speed and affordability. This system card addendum provides updates on the baseline safety metrics and evaluation results for GPT-6.1 Sol. Under our Preparedness Framework, we are treating GPT-6.1 Sol as Critical in cybersecurity and High for Biological and Chemical capability. Based on that assessment, GPT-6.1 Sol uses the same safeguards stack as GPT-6 Astra, described in detail in the GPT-6 Astra system card. Evaluations of GPT-6.1 Sol were performed in our research environment or via our API, which may provide slightly different output from production ChatGPT due to differences in the system prompts, tools available, efforts etc. 2. Model Data and Training GPT-6.1 Sol uses the same types of data and training as GPT-6 Astra, described in the GPT-6 Astra card. GPT-6.1 Sol is intended to be used in accordance with OpenAI’s Usage Policies, Service Terms, and Terms of Use. These policies apply universally to OpenAI services and are designed to ensure safe and responsible usage of AI technology. You can review OpenAI’s Usage Policies at openai.com/policies/usage-policies/ . 3. Model Safety For a full description of the evaluations in this section, please see the Model Safety section in the GPT-6 Astra card. 3.1 Safe Completions We conducted benchmark evaluations across safety categories. We report here on our Production Benchmarks, an evaluation set with conversations representative of challenging examples from production data. On our Production Benchmarks, GPT-6.1 Sol scores higher than GPT-6 Sol in five of eight categories. Our evaluations indicate that our GPT-6 models are less likely to refuse harmless requests or add excessive or unnecessarily judgmental caveats. GPT‑6.1 Sol scores higher than GPT‑6 Sol in five of six under-18 safety categories; the regression on the gore evaluation is not statistically significant. We find that GPT-6.1 Sol is generally better than GPT-6 Sol at recognizing potential harms in agentic requests and avoiding taking harmful actions. 3.2 Vision Across all vision evaluations, GPT-6.1 Sol performed on par with or better than GPT‑6 Sol. 4. Robustness On both the static and multiturn jailbreak evaluations, GPT-6.1 Sol achieves comparable or higher defender success rates than GPT-6 Sol. All models in the GPT-6 series show an improvement over earlier models. GPT-6 Astra, GPT-6.1 Sol, and GPT-6 Sol are the most robust frontier models across both our static and multiturn robustness evaluations. We find that GPT-6 Sol and GPT-6 Luna have substantial improvements to prompt injection and instruction hierarchy robustness compared to their predecessors. GPT-6.1 Sol is also highly robust. 5. Health GPT-6.1 Sol performs on par with GPT-6 Astra on HealthBench evaluations, with length-adjusted scores within 0.5 percentage points of GPT‑6 Astra across HealthBench Professional, HealthBench, HealthBench Hard, and HealthBench Consensus. 9. Preparedness Under our Preparedness Framework, we are treating GPT-6.1 Sol as Critical in cybersecurity and High for Biological and Chemical capability. Based on that assessment, GPT-6.1 Sol uses the same safeguards stack as GPT-6 Astra. On the cybersecurity safety evaluations, GPT-6.1 Sol outperforms all our previous models in production-chat evaluations. Compared to GPT-5.6 Sol, GPT-6.1 Sol shows modest regressions in synthetic and semi-synthetic agentic environments. As cyber capabilities increase, our approach is to expand defensive access and safeguards together. Like Astra, we are taking a phased approach for GPT-6.1 Sol through Daybreak and, over time, our goal is to enable more advanced, authorized work through more precise safeguards, supported by stronger verification, accountability, and monitoring. The full system card addendum, including evaluation tables and figures, is published on OpenAI’s Deployment Safety Hub: https://deploymentsafety.openai.com/gpt-6-1-sol .

## Key details

- Issued by: OpenAI
- Published: 29 Sep 2026
- Publisher country: United States
- Subject country/region: United States
- Topics: AI
- Original source: https://deploymentsafety.openai.com/gpt-6-1-sol
- MediaRelease.co URL: https://mediarelease.co/us/mr01097-addendum-gpt-6-astra-system-card-gpt-6-1-sol-29092026.html

Original source: https://deploymentsafety.openai.com/gpt-6-1-sol
MediaRelease.co canonical URL: https://mediarelease.co/us/mr01097-addendum-gpt-6-astra-system-card-gpt-6-1-sol-29092026.html