Search results for A9
A9 community
One keyword maps to one global community path.
People
Not Found
Tweets including A9
Qwen3.8-2.4T-A95B, compressed two ways at once: 25% of the experts pruned with REAP, and the rest quantized to NVFP4.
Even with a quarter of the experts gone and 4-bit weights, GPQA Diamond holds at 91.5 vs 92.6 for the full-precision base. ~99% recovery.
Serve on @vllm_project:
Show more
Qwen3.8-2.4T-A95B, a 2.4-trillion parameter MoE, quantized with no accuracy loss.
MoE layers to NVFP4, attention layers to FP8 block, via LLM Compressor. On GPQA Diamond: 93.1 quantized vs 92.6 for the full-precision base. Full recovery.
Serve on vLLM:
Show more
Qwen3.8-2.4T-A95B is live on Together AI.
The Qwen team’s new open-weight MoE packs 2.4T total parameters with 95B active, a 1M context window, and strong coding & agent capabilities.
Start building:
Show more
