Comparing
compact-bytecode-encoding-with-fused-opcodes
(64b3dbc
) withcanary
(3f888bf
)+34%
- Improvement
- 9
- Untouched
- 10
Unknown Walltime execution environment detected
Using the Walltime instrument on standard Hosted Runners will lead to inconsistent data. For the most accurate results, we recommend using CodSpeed Macro Runners: bare-metal machines fine-tuned for performance measurement consistency.Benchmarks
19 total
vm_class_create_50k
crates/baml_tests/benches/runtime_benchmark.rs+34%
19.3 ms14.4 ms
vm_field_access_50k
crates/baml_tests/benches/runtime_benchmark.rs+27%
9.7 ms7.6 ms
vm_loop_500k
crates/baml_tests/benches/runtime_benchmark.rs+27%
45.4 ms35.7 ms
vm_array_iter_10k
crates/baml_tests/benches/runtime_benchmark.rs+22%
5.8 ms4.7 ms
vm_array_push_50k
crates/baml_tests/benches/runtime_benchmark.rs+19%
9.7 ms8.2 ms
vm_closure_call_50k
crates/baml_tests/benches/runtime_benchmark.rs+17%
10.7 ms9.1 ms
vm_fib_20
crates/baml_tests/benches/runtime_benchmark.rs+17%
5.5 ms4.7 ms
vm_nested_loop
crates/baml_tests/benches/runtime_benchmark.rs+14%
6.1 ms5.4 ms
vm_call_chain_100_x_5k
crates/baml_tests/benches/runtime_benchmark.rs+11%
36.6 ms33.1 ms
engine_init_cost
crates/baml_tests/benches/runtime_benchmark.rs+10%
2.3 ms2.1 ms
startup_empty_expression
crates/baml_tests/benches/runtime_benchmark.rs+3%
42.6 ms41.4 ms
vm_mixed_ops
crates/baml_tests/benches/runtime_benchmark.rs+2%
19 ms18.5 ms
e2e_hello_world
crates/baml_tests/benches/runtime_benchmark.rs+1%
41.8 ms41.5 ms
e2e_fib_20
crates/baml_tests/benches/runtime_benchmark.rs0%
45.2 ms45.2 ms
e2e_class_and_loop
crates/baml_tests/benches/runtime_benchmark.rs0%
42.9 ms42.9 ms
e2e_arithmetic
crates/baml_tests/benches/runtime_benchmark.rs0%
41.6 ms41.7 ms
e2e_100_functions
crates/baml_tests/benches/runtime_benchmark.rs-1%
100.1 ms100.7 ms
compile_to_engine
crates/baml_tests/benches/runtime_benchmark.rs-1%
41.2 ms41.7 ms
vm_string_concat_5k
crates/baml_tests/benches/runtime_benchmark.rs-6%
44.9 ms47.6 ms
Commits
Click on a commit to change the comparison rangeBase
canary
3f888bf
+27.63%
perf: tight inner dispatch loop for compact bytecode (1.0-1.2x faster)
1a9b03a
16 days ago
by hellovai-45.88%
fix: lower bytecode to compact in convert_program for non-engine callers
9df9dee
16 days ago
by hellovai+52.41%
fix: compact disassembler reads correct number of MakeClosure operands
64b3dbc
16 days ago
by hellovai