oneapi-agent-toolkit

Model Example Arm correct + 0 helpers wrong output dpct helpers left said done but wasn’t cost
claude-haiku-5-5 all toolkit 20/20 0/20 0/20 0/20 $0.08
claude-haiku-5-5 all baseline 13/20 4/20 5/20 7/20 $0.07
claude-haiku-5-5 all baseline_shell 7/20 0/20 13/20 13/20 $0.10
claude-haiku-5-5 saxpy toolkit 5/5 0/5 0/5 0/5 $0.01
claude-haiku-5-5 saxpy baseline 0/5 3/5 3/5 5/5 $0.01
claude-haiku-5-5 saxpy baseline_shell 0/5 0/5 5/5 5/5 $0.02
claude-haiku-5-5 fuzz toolkit 5/5 0/5 0/5 0/5 $0.01
claude-haiku-5-5 fuzz baseline 5/5 0/5 0/5 0/5 $0.01
claude-haiku-5-5 fuzz baseline_shell 1/5 0/5 4/5 4/5 $0.02
claude-haiku-5-5 fir toolkit 5/5 0/5 0/5 0/5 $0.03
claude-haiku-5-5 fir baseline 3/5 1/5 2/5 2/5 $0.02
claude-haiku-5-5 fir baseline_shell 3/5 0/5 2/5 2/5 $0.03
claude-haiku-5-5 softmax_hecbench toolkit 5/5 0/5 0/5 0/5 $0.03
claude-haiku-5-5 softmax_hecbench baseline 5/5 0/5 0/5 0/5 $0.03
claude-haiku-5-5 softmax_hecbench baseline_shell 3/5 0/5 2/5 2/5 $0.04
claude-sonnet-5-5 all toolkit 20/20 0/20 0/20 0/20 $1.37
claude-sonnet-5-5 all baseline 10/20 2/20 10/20 10/20 $1.23
claude-sonnet-5-5 all baseline_shell 7/20 0/20 13/20 12/20 $1.20
claude-sonnet-5-5 saxpy toolkit 5/5 0/5 0/5 0/5 $0.17
claude-sonnet-5-5 saxpy baseline 1/5 0/5 4/5 4/5 $0.30
claude-sonnet-5-5 saxpy baseline_shell 3/5 0/5 2/5 2/5 $0.14
claude-sonnet-5-5 fuzz toolkit 5/5 0/5 0/5 0/5 $0.43
claude-sonnet-5-5 fuzz baseline 4/5 0/5 1/5 1/5 $0.38
claude-sonnet-5-5 fuzz baseline_shell 1/5 0/5 4/5 4/5 $0.35
claude-sonnet-5-5 fir toolkit 5/5 0/5 0/5 0/5 $0.30
claude-sonnet-5-5 fir baseline 3/5 0/5 2/5 2/5 $0.24
claude-sonnet-5-5 fir baseline_shell 2/5 0/5 3/5 3/5 $0.29
claude-sonnet-5-5 softmax_hecbench toolkit 5/5 0/5 0/5 0/5 $0.48
claude-sonnet-5-5 softmax_hecbench baseline 2/5 2/5 3/5 3/5 $0.31
claude-sonnet-5-5 softmax_hecbench baseline_shell 1/5 0/5 4/5 3/5 $0.43