Opus and Qwen do a Sokoban

In a head-to-head coding test, Claude Opus 5.5 scored 77% and locally run Qwen3.8-Flash-Next scored 70% across 30 implementation aspects of a single-file HTML/JS/CSS Sokoban game, according to blogger Senko's vibecode benchmarks. Opus finished the one-line prompt in about 4 minutes, while Qwen — quantized to 4 bits on an AMD Strix Halo machine with 128GB of unified RAM — took about 31 minutes and 64k tokens, roughly 80% of it reasoning at extra-high effort. GPT-6 Astra, used as an unbiased judge, concluded Opus held "a modest but meaningful advantage in correctness and maintainability," and Senko noted the local setup costs $3.5k to $6k, comparable to a few years of Claude Max.

Opus and Qwen do a Sokoban Comparing Claude Opus 5.5 vs Qwen3.8-Flash-Next, running locally, on the same coding task: Sokoban https://en.wikipedia.org/wiki/Sokoban In a slightly different setup than usual, I ran Claude Opus 5.5 and Qwen3.8-Flash-Next through the chatbot interface not a coding agent ; Qwen I ran via llama.app https://llama.app/ quantized to 4 bits https://unsloth.ai/docs/models/qwen3.8-next on an AMD Strix Halo machine with 128GB of unified RAM . The prompt was a one-liner: “ Create a simple but beautiful Sokoban game clone in HTML/JS/CSS all in one file, no frameworks or 3rd-party libraries . “ Opus high reasoning effort took ~4 minutes I don't have a token count because this was from a web chatbot interface . Qwen took ~31 minutes and 64k tokens. 80% of this was for reasoning at extra-high effort . Qwen thinks a lot more than Opus Both are available at my vibecode benchmarks page https://senko.net/vibecode-bench/ . Here's also a side-by-side playthrough of both: I then had GPT-6 Astra compare the quality of the implementations. I chose Astra so as to not have any bias towards a solution in case models really do prefer their own . Across 30 different areas/aspects of each solution, Opus scored 77% and Qwen 70% 100% would be a perfect implementation without downsides or bugs . Astra concluded the code review with: “ Claude’s implementation has a modest but meaningful advantage in correctness and maintainability, with simpler state management, stronger movement safeguards, and fewer consequential bugs. “ Note that this test compared the best frontier model for coding against something I ran completely locally, on a machine sitting next to me. As I wrote earlier this year, open-weights models have become good enough https://blog.senko.net/open-weights-ai-models-have-become-good-enough , but it's hard and expensive to run a very capable model locally. For this test I chose Sokoban because, frankly, my usual “frontier” tests, RTS/Warcraft, Minecraft and a flying simulator, were beyond Qwen's reach for a zero-shot. When trying to build an RTS, the model took hours and produced an unusable mess. The hardware cost for this setup can vary between $3.5k and $6k, depending on luck, availability, and other system components. Not cheap, but not beyond the budget for a small business or affluent techies. The same amount of money does buy you a few years of Claude Max subscription—purely on that level, it's not easy to justify. But while I will not be canceling my Claude subscription any time soon, I'd say that for smaller tasks and with more guided steering, this is a usable alternative for those who would prefer to—or must—use local models.