github turboderp-org/exllamav3 v0.0.19
0.0.19

latest releases: v1.6.1, v1.6.0, v1.5.4...
9 months ago
  • Support Olmo3ForCausalLM
  • Support HyperCLOVAXForCausalLM (and HCXVisionV2ForCausalLM)
  • Support SolarOpenForCausalLM
  • Support NanoChatForCausalLM
  • Better support for quantizing from FP8 weights
  • Fix memory leak for large job queues on Qwen3-Next
  • Add Adaptive-P sampler
  • QoL improvements and bugfixes

Full Changelog: v0.0.18...v0.0.19

Don't miss a new exllamav3 release

NewReleases is sending notifications on new releases.