Benchmark result

Swift-1.5-IQ3_XSS on AMD Radeon RX 7800 XT +1 more — 48.9 tok/s

Measured with OpenAI-compatible manual on November 6, 2026.

ShareRedditX

Support llm-bench.io

llm-bench.io is free. If it helped you, I'd appreciate your support to cover the infrastructure costs.

Support me on Ko-fiBuy me a coffee

What the model built

Open full screen →

The coding scenario asks for a playable game in a single HTML file. This is exactly what the model returned, unedited; the frame only adds a line that tells this page how tall it is.

Reproduce this run

toolOpenAI-compatible manual
modelSwift-1.5-IQ3_XSS
context128,000
thinkingon
clientv0.4.98+102

Set those in OpenAI-compatible, then:

llm-benchmark benchmark --model "Swift-1.5-IQ3_XSS" --tool "OpenAI-compatible"

The client prompts for context length and thinking mode, and for the KV cache dtype on Unsloth Studio. Sampling is left at the model default — the values above are what the tool reported using, not overrides the client sent.