OpenAI benchmark: GPT-6.1 Sol hits 97% accuracy migrating COBOL at $0.50 per trial
ChrisGPT · x · 2026-10-20
OpenAI's new COBOL migration benchmark shows its models are exceptionally good at modernizing legacy software that still powers much of the global financial services industry.
- GPT-6.1 Sol: 97% accuracy at roughly $0.50 per graded trial
- GPT-6 Sol: needs about $2 per trial to reach similar performance
- The shared chart also compares GPT-5.6 Sol and GPT-6 Astra, with the newest generation dominating on both cost and accuracy
The results suggest AI-driven automation of large-scale legacy system migration is becoming rapidly more viable.
More from coding & agent
- OpenCode hacked: video surfaces on security breach of open-source coding tool — RelevantEmergency707 · 2026-10-20
- Stanford researchers study cascade harnesses to maximize agent performance at low cost — james_y_zou · 2026-10-20
- Dev ran a Shopify store end-to-end with Hermes-Agent, making physical merch — alexcovo_eth · 2026-10-20
- LlamaIndex launches ExtractBench Pro: long-tail document extraction that stumps frontier models — llama_index · 2026-10-20
- vLLM ecosystem ships d3-turbo: multimodal decision model, 15x faster, tops Jev Decision Index — vllm_project · 2026-10-20
- Compacting a compacted LLM context: engineers' silent pain — nrehiew_ · 2026-10-20