feat(server): Using quantize_config.json
instead of GPTQ_BITS env variables.
#1054
Job | Run time |
---|---|
1m 35s | |
5m 34s | |
15m 18s | |
5s | |
22m 32s |
quantize_config.json
instead of GPTQ_BITS env variables.
#1054
Job | Run time |
---|---|
1m 35s | |
5m 34s | |
15m 18s | |
5s | |
22m 32s |