Open-Source LLM and Leaderboard 2026 Collaboration Announced
benchmarks claude deepseek llama open-source qwen
| Source: Mastodon | Original article
Open-source LLMs ranked: Qwen3.8 Max trails Claude Opus 5. Qwen3.8 is 4x cheaper per output token.
The Open-Source LLM Leaderboard 2026 has been released, providing a comprehensive comparison of open-source and open-weight large language models. Qwen3.8 Max is currently trailing Claude Opus 5 by 4.5 points, but its significantly lower cost of 4x cheaper per 1M output tokens narrows the practical gap between the two models.
This leaderboard matters as it highlights the growing competitiveness of open-source LLMs, offering enterprise-grade performance without the costs and vendor lock-in associated with proprietary systems. The rankings also consider factors such as pricing, speed, and context windows, making it a valuable resource for practitioners and researchers.
As the LLM landscape continues to evolve, this leaderboard will be an important tool for tracking the performance and progress of open-source models. With the ability to filter and share custom views, users can stay up-to-date on the latest developments and find the best model for their specific needs. The open-source community will likely be watching closely to see how these models continue to improve and innovate in the coming months.
Sources
Back to AIPULSEN