ZAYA1-8B: An 8B Moe Model with 760M Active Params Matching DeepSeek-R1 on Math
ZAYA1-8B matches DeepSeek-R1 on math benchmarks and remains competitive with Claude Sonnet 4.5 on reasoning tasks. The model, trained entirely on AMD hardware, operates with less than 1 billion active parameters while closing in on Gemini 2.5 Pro in coding performance.
firethering.com
6 min
5/7/2026
Qwen3.6-35B-A3B on my laptop drew me a better pelican than Claude Opus 4.7
Qwen3.6-35B-A3B generated a superior image of a pelican compared to Claude Opus 4.7. The Qwen model was run on a MacBook Pro M5 using LM Studio.
simonwillison.net
2 min
4/16/2026
ZAYA1-8B: An 8B Moe Model with 760M Active Params Matching DeepSeek-R1 on Math
ZAYA1-8B matches DeepSeek-R1 on math benchmarks and remains competitive with Claude Sonnet 4.5 on reasoning tasks. The model, trained entirely on AMD hardware, operates with less than 1 billion active parameters while closing in on Gemini 2.5 Pro in coding performance.
firethering.com
6 min
5/7/2026
Qwen3.6-35B-A3B on my laptop drew me a better pelican than Claude Opus 4.7
Qwen3.6-35B-A3B generated a superior image of a pelican compared to Claude Opus 4.7. The Qwen model was run on a MacBook Pro M5 using LM Studio.
simonwillison.net
2 min
4/16/2026
ZAYA1-8B: An 8B Moe Model with 760M Active Params Matching DeepSeek-R1 on Math
ZAYA1-8B matches DeepSeek-R1 on math benchmarks and remains competitive with Claude Sonnet 4.5 on reasoning tasks. The model, trained entirely on AMD hardware, operates with less than 1 billion active parameters while closing in on Gemini 2.5 Pro in coding performance.
firethering.com
6 min
5/7/2026
Qwen3.6-35B-A3B on my laptop drew me a better pelican than Claude Opus 4.7
Qwen3.6-35B-A3B generated a superior image of a pelican compared to Claude Opus 4.7. The Qwen model was run on a MacBook Pro M5 using LM Studio.
simonwillison.net
2 min
4/16/2026
No more articles to load