mlx-bench c51b247: bench_mistral regression test

bench_mistral/Apple_M1_Max/ml-explore-mlx@[email protected]

@@ -1 +1 @@
-{"args": {"hub_url": "github.com", "repo_a": "ml-explore/mlx", "repo_b": "ml-explore/mlx", "commit_a": "cbcf44a4caf3fb504ed29ef78091126134e197a3", "commit_b": "14b4e51a7c6455a61a74d24da9f47dfeb161023f", "output_dir": "external", "hub_model_name": "mlx-community/Mistral-7B-Instruct-v0.2-4-bit", "max_context_length": 800, "measure_every_n_tokens": 100, "fail_for_mismatch_before_n_tokens": 800, "bench_cmd": "python llms/mistral/mistral.py --model-path ../external/model --max-tokens 800 --tokens-per-eval 100 --prompt 'Continue this series forever: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10' --metal-disallow-cache"}, "repo_a": [{"kv_cache_length": 99, "tokens_per_sec": 27.4, "generated_text": ", 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35", "peak_memory": 4338623610, "cache_memory": 0, "active_memory": 4333642284}, {"kv_cache_length": 199, "tokens_per_sec": 25.37, "generated_text": ", 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60", "peak_memory": 4394364682, "cache_memory": 0, "active_memory": 4386071084}, {"kv_cache_length": 299, "tokens_per_sec": 23.74, "generated_text": ", 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85", "peak_memory": 4450206654, "cache_memory": 0, "active_memory": 4438499884}, {"kv_cache_length": 399, "tokens_per_sec": 24.27, "generated_text": ", 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 10", "peak_memory": 4505977970, "cache_memory": 0, "active_memory": 4490928684}, {"kv_cache_length": 499, "tokens_per_sec": 24.51, "generated_text": "8, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 12", "peak_memory": 4561814260, "cache_memory": 0, "active_memory": 4543357484}, {"kv_cache_length": 599, "tokens_per_sec": 23.13, "generated_text": "8, 129, 130, 131, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 14", "peak_memory": 4617520040, "cache_memory": 0, "active_memory": 4595786284}, {"kv_cache_length": 699, "tokens_per_sec": 22.33, "generated_text": "8, 149, 150, 151, 152, 153, 154, 155, 156, 157, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 16", "peak_memory": 4673291378, "cache_memory": 0, "active_memory": 4648215084}, {"kv_cache_length": 799, "tokens_per_sec": 22.13, "generated_text": "8, 169, 170, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 181, 182, 183, 184, 185, 186, 187, 18", "peak_memory": 4729127668, "cache_memory": 0, "active_memory": 4700643884}], "repo_b": [{"kv_cache_length": 99, "tokens_per_sec": 42.69, "generated_text": ", 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35", "peak_memory": 4288671016, "cache_memory": 0, "active_memory": 4280164908}, {"kv_cache_length": 199, "tokens_per_sec": 41.79, "generated_text": ", 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60", "peak_memory": 4296375184, "cache_memory": 0, "active_memory": 4293796396}, {"kv_cache_length": 299, "tokens_per_sec": 40.5, "generated_text": ", 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85", "peak_memory": 4309809868, "cache_memory": 0, "active_memory": 4306379308}, {"kv_cache_length": 399, "tokens_per_sec": 40.31, "generated_text": ", 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 10", "peak_memory": 4324296252, "cache_memory": 0, "active_memory": 4320010796}, {"kv_cache_length": 499, "tokens_per_sec": 39.08, "generated_text": "8, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 12", "peak_memory": 4337763404, "cache_memory": 0, "active_memory": 4332593708}, {"kv_cache_length": 599, "tokens_per_sec": 37.37, "generated_text": "8, 129, 130, 131, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 14", "peak_memory": 4352248844, "cache_memory": 0, "active_memory": 4346225196}, {"kv_cache_length": 699, "tokens_per_sec": 24.94, "generated_text": "8, 149, 150, 151, 152, 153, 154, 155, 156, 157, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 16", "peak_memory": 4365717028, "cache_memory": 0, "active_memory": 4358808108}, {"kv_cache_length": 799, "tokens_per_sec": 36.42, "generated_text": "8, 169, 170, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 181, 182, 183, 184, 185, 186, 187, 18", "peak_memory": 4380250012, "cache_memory": 0, "active_memory": 4372439596}], "mismatch_after_n_tokens": null, "inference_ctx": {"os_spec": {"os_version": "14.1", "os_type": "macOS", "os_build_number": "23B73"}, "code_spec": {"repo_a": "ml-explore/mlx", "repo_b": "ml-explore/mlx", "commit_a": "cbcf44a4caf3fb504ed29ef78091126134e197a3", "commit_b": "14b4e51a7c6455a61a74d24da9f47dfeb161023f"}, "model_spec": {"hub_model_name": "mlx-community/Mistral-7B-Instruct-v0.2-4-bit"}, "device_spec": {"gpu_core_count": 32, "cpu_core_count": 10, "product_name": "Apple M1 Max", "max_ram": "67794468864"}}, "mlx_bench_commit": "c51b247"}
+{"args": {"hub_url": "github.com", "repo_a": "ml-explore/mlx", "repo_b": "ml-explore/mlx", "commit_a": "cbcf44a4caf3fb504ed29ef78091126134e197a3", "commit_b": "14b4e51a7c6455a61a74d24da9f47dfeb161023f", "output_dir": "external", "hub_model_name": "mlx-community/Mistral-7B-Instruct-v0.2-4-bit", "max_context_length": 800, "measure_every_n_tokens": 100, "fail_for_mismatch_before_n_tokens": 800, "bench_cmd": "python llms/mistral/mistral.py --model-path ../external/model --max-tokens 800 --tokens-per-eval 100 --prompt 'Continue this series forever: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10'"}, "repo_a": [{"kv_cache_length": 99, "tokens_per_sec": 30.96, "generated_text": ", 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35", "peak_memory": 4338664870, "cache_memory": 2349169391, "active_memory": 4333658668}, {"kv_cache_length": 199, "tokens_per_sec": 29.17, "generated_text": ", 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60", "peak_memory": 4394370546, "cache_memory": 7505214783, "active_memory": 4386087468}, {"kv_cache_length": 299, "tokens_per_sec": 24.52, "generated_text": ", 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85", "peak_memory": 4450207218, "cache_memory": 15442001215, "active_memory": 4438516268}, {"kv_cache_length": 399, "tokens_per_sec": 15.91, "generated_text": ", 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 10", "peak_memory": 4506043710, "cache_memory": 26168606015, "active_memory": 4490945068}, {"kv_cache_length": 499, "tokens_per_sec": 19.24, "generated_text": "8, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 12", "peak_memory": 4561814846, "cache_memory": 39680490815, "active_memory": 4543373868}, {"kv_cache_length": 599, "tokens_per_sec": 19.17, "generated_text": "8, 129, 130, 131, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 14", "peak_memory": 4617520424, "cache_memory": 44365062328, "active_memory": 4595802668}, {"kv_cache_length": 699, "tokens_per_sec": 20.26, "generated_text": "8, 149, 150, 151, 152, 153, 154, 155, 156, 157, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 16", "peak_memory": 4673291380, "cache_memory": 44311453880, "active_memory": 4648231468}, {"kv_cache_length": 799, "tokens_per_sec": 20.73, "generated_text": "8, 169, 170, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 181, 182, 183, 184, 185, 186, 187, 18", "peak_memory": 4729193948, "cache_memory": 44258304184, "active_memory": 4700660268}], "repo_b": [{"kv_cache_length": 99, "tokens_per_sec": 45.97, "generated_text": ", 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35", "peak_memory": 4289998276, "cache_memory": 164848043, "active_memory": 4280181292}, {"kv_cache_length": 199, "tokens_per_sec": 43.16, "generated_text": ", 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60", "peak_memory": 4296409802, "cache_memory": 483666887, "active_memory": 4293812780}, {"kv_cache_length": 299, "tokens_per_sec": 42.26, "generated_text": ", 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85", "peak_memory": 4309815379, "cache_memory": 944717511, "active_memory": 4306395692}, {"kv_cache_length": 399, "tokens_per_sec": 41.26, "generated_text": ", 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 10", "peak_memory": 4324329747, "cache_memory": 1640778055, "active_memory": 4320027180}, {"kv_cache_length": 499, "tokens_per_sec": 40.38, "generated_text": "8, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 12", "peak_memory": 4337767875, "cache_memory": 2437735111, "active_memory": 4332610092}, {"kv_cache_length": 599, "tokens_per_sec": 38.88, "generated_text": "8, 129, 130, 131, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 14", "peak_memory": 4352282399, "cache_memory": 3495883207, "active_memory": 4346241580}, {"kv_cache_length": 699, "tokens_per_sec": 38.38, "generated_text": "8, 149, 150, 151, 152, 153, 154, 155, 156, 157, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 16", "peak_memory": 4365721322, "cache_memory": 4639758535, "active_memory": 4358824492}, {"kv_cache_length": 799, "tokens_per_sec": 37.61, "generated_text": "8, 169, 170, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 181, 182, 183, 184, 185, 186, 187, 18", "peak_memory": 4380268842, "cache_memory": 6059994695, "active_memory": 4372455980}], "mismatch_after_n_tokens": null, "inference_ctx": {"os_spec": {"os_version": "14.1", "os_type": "macOS", "os_build_number": "23B73"}, "code_spec": {"repo_a": "ml-explore/mlx", "repo_b": "ml-explore/mlx", "commit_a": "cbcf44a4caf3fb504ed29ef78091126134e197a3", "commit_b": "14b4e51a7c6455a61a74d24da9f47dfeb161023f"}, "model_spec": {"hub_model_name": "mlx-community/Mistral-7B-Instruct-v0.2-4-bit"}, "device_spec": {"gpu_core_count": 32, "cpu_core_count": 10, "product_name": "Apple M1 Max", "max_ram": "67794468864"}}, "mlx_bench_commit": "c51b247"}