Download scripts/tbench/gemma_env.sh from jaytonde05/kv-cache-quantization-commands: direct link, hf CLI and curl.
- Browser
- Download file 473 Bytes
-
https://huggingface.co/datasets/jaytonde05/kv-cache-quantization-commands/resolve/main/scripts/tbench/gemma_env.sh
- Command line
-
hf download hf://datasets/jaytonde05/kv-cache-quantization-commands/scripts/tbench/gemma_env.sh
-
curl -L -o gemma_env.sh https://huggingface.co/datasets/jaytonde05/kv-cache-quantization-commands/resolve/main/scripts/tbench/gemma_env.sh
473 Bytes
| # Gemma 4 31B settings for tb_queue.sh / tb_server.sh (source before running). Thinking ON (model-card setting). | |
| # Thinking-off variant (2026-10-25 run, saved in ~/tbench/jobs/gemma_thinking_off): same with "enable_thinking": false. | |
| export KVQ_MODEL=~/models/gemma-4-31B-it KVQ_TB_NAME=gemma-4-31b-it KVQ_REASONING_PARSER=gemma4 KVQ_TB_EFFORT= \ | |
| KVQ_TB_LOG_PREFIX=gemma- KVQ_SERVE_ARGS="--language-model-only --default-chat-template-kwargs '{\"enable_thinking\": true}'" | |