{
  "date": "2026-09-12T17:55:31.667660+00:00",
  "gpu": "NVIDIA GeForce RTX 5070 Ti Laptop GPU, 12227 MiB, 610.43.02",
  "cpu": "Architecture\u00a0:                            x86_64\nMode(s) op\u00e9ratoire(s) des processeurs\u00a0:   32-bit, 64-bit\nTailles des adresses:                     42 bits physical, 48 bits virtual\nBoutisme\u00a0:                                Little Endian\nProcesseur(s)\u00a0:                           24\nListe de processeur(s) en ligne\u00a0:         0-23\nIdentifiant constructeur\u00a0:                GenuineIntel\nNom de mod\u00e8le\u00a0:                           Intel(R) Core(TM) Ultra 9 275HX\nFamille de processeur\u00a0:                   6\nMod\u00e8le\u00a0:                                  198\nThread(s) par c\u0153ur\u00a0:                      1\nC\u0153ur(s) par socket\u00a0:                      24\nSocket(s)\u00a0:                               1\nR\u00e9vision\u00a0:                                2\nVersion du microcode\u00a0:                    0x121\nmultiplication des MHz du/des CPU(s)\u00a0:    34%\nVitesse maximale du processeur en MHz\u00a0:   5400,0000\nVitesse minimale du processeur en MHz\u00a0:   800,0000\nBogoMIPS\u00a0:                                6144,00\nDrapeaux\u00a0:                                fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc art arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf tsc_known_freq pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm 3dnowprefetch cpuid_fault ssbd ibrs ibpb stibp ibrs_enhanced tpr_shadow flexpriority ept vpid ept_ad fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid rdt_a rdseed adx smap clflushopt clwb intel_pt sha_ni xsaveopt xsavec xgetbv1 xsaves split_lock_detect user_shstk avx_vnni lam wbnoinvd dtherm ida arat pln pts hwp hwp_notify hwp_act_window hwp_epp hwp_pkg_req hfi vnmi umip pku ospke waitpkg gfni vaes vpclmulqdq rdpid bus_lock_detect movdiri movdir64b fsrm md_clear serialize arch_lbr ibt flush_l1d arch_capabilities\nVirtualisation\u00a0:                          VT-x\nCache L1d\u00a0:                               768 KiB (20 instances)\nCache L1i\u00a0:                               1,3 MiB (20 instances)\nCache L2\u00a0:                                40 MiB (12 instances)\nCache L3\u00a0:                                36 MiB (1 instance)\nN\u0153ud(s) NUMA\u00a0:                            1\nN\u0153ud NUMA\u00a00 de processeur(s)\u00a0:            0-23\nVuln\u00e9rabilit\u00e9 Gather data sampling\u00a0:      Not affected\nVuln\u00e9rabilit\u00e9 Ghostwrite\u00a0:                Not affected\nVuln\u00e9rabilit\u00e9 Indirect target selection\u00a0: Not affected\nVuln\u00e9rabilit\u00e9 Itlb multihit\u00a0:             Not affected\nVuln\u00e9rabilit\u00e9 L1tf\u00a0:                      Not affected\nVuln\u00e9rabilit\u00e9 Mds\u00a0:                       Not affected\nVuln\u00e9rabilit\u00e9 Meltdown\u00a0:                  Not affected\nVuln\u00e9rabilit\u00e9 Mmio stale data\u00a0:           Not affected\nVuln\u00e9rabilit\u00e9 Old microcode\u00a0:             Not affected\nVuln\u00e9rabilit\u00e9 Reg file data sampling\u00a0:    Not affected\nVuln\u00e9rabilit\u00e9 Retbleed\u00a0:                  Not affected\nVuln\u00e9rabilit\u00e9 Spec rstack overflow\u00a0:      Not affected\nVuln\u00e9rabilit\u00e9 Spec store bypass\u00a0:         Mitigation; Speculative Store Bypass disabled via prctl\nVuln\u00e9rabilit\u00e9 Spectre v1\u00a0:                Mitigation; usercopy/swapgs barriers and __user pointer sanitization\nVuln\u00e9rabilit\u00e9 Spectre v2\u00a0:                Mitigation; Enhanced / Automatic IBRS; IBPB conditional; PBRSB-eIBRS Not affected; BHI BHI_DIS_S\nVuln\u00e9rabilit\u00e9 Srbds\u00a0:                     Not affected\nVuln\u00e9rabilit\u00e9 Tsa\u00a0:                       Not affected\nVuln\u00e9rabilit\u00e9 Tsx async abort\u00a0:           Not affected\nVuln\u00e9rabilit\u00e9 Vmscape\u00a0:                   Mitigation; IBPB before exit to userspace",
  "os": "Linux cachyos-mgm 6.18.33-2-cachyos-lts #1 SMP PREEMPT_DYNAMIC Thu, 28 May 2026 22:25:50 +0000 x86_64 GNU/Linux",
  "runtime": {
    "version": "0.30.6"
  },
  "models": {
    "models": [
      {
        "name": "glm-4.7-flash:latest",
        "model": "glm-4.7-flash:latest",
        "modified_at": "2026-08-26T11:39:31.967610161+02:00",
        "size": 19019270897,
        "digest": "4475827791a269b02c8ec49b1c3bc1abb5846bacf3fae015b75d33986322d8f6",
        "details": {
          "parent_model": "glm-4.7-flash:latest",
          "format": "gguf",
          "family": "glm4moelite",
          "families": [
            "glm4moelite"
          ],
          "parameter_size": "29.9B",
          "quantization_level": "Q4_K_M",
          "context_length": 202752,
          "embedding_length": 2048
        },
        "capabilities": [
          "completion",
          "tools",
          "thinking"
        ]
      },
      {
        "name": "gemma4:e4b-it-qat",
        "model": "gemma4:e4b-it-qat",
        "modified_at": "2026-06-07T23:19:34.645395154+02:00",
        "size": 6146501801,
        "digest": "ee665637121887cf3befff38abbb1be4ee117c7db867d97a67e29049ecd7e15f",
        "details": {
          "parent_model": "",
          "format": "gguf",
          "family": "gemma4",
          "families": [
            "gemma4"
          ],
          "parameter_size": "7.5B",
          "quantization_level": "Q4_0",
          "context_length": 131072,
          "embedding_length": 2560
        },
        "capabilities": [
          "completion",
          "tools",
          "thinking",
          "vision"
        ]
      },
      {
        "name": "gemma4:12b-it-qat",
        "model": "gemma4:12b-it-qat",
        "modified_at": "2026-06-07T23:18:33.914062685+02:00",
        "size": 7151003754,
        "digest": "38044be4f923e5a55264ed7df4eaac2676651a905f735197c504045140c02bd3",
        "details": {
          "parent_model": "",
          "format": "gguf",
          "family": "gemma4",
          "families": [
            "gemma4"
          ],
          "parameter_size": "11.9B",
          "quantization_level": "Q4_0",
          "context_length": 262144,
          "embedding_length": 3840
        },
        "capabilities": [
          "completion",
          "tools",
          "thinking",
          "vision"
        ]
      },
      {
        "name": "mistral-nemo:12b",
        "model": "mistral-nemo:12b",
        "modified_at": "2026-06-07T22:54:08.052027827+02:00",
        "size": 7071713227,
        "digest": "e7e06d107c6c86ed0cf45445f1790720b5092149c4c95f4d965844e9afbfdc89",
        "details": {
          "parent_model": "",
          "format": "gguf",
          "family": "llama",
          "families": [
            "llama"
          ],
          "parameter_size": "12.2B",
          "quantization_level": "Q4_0",
          "context_length": 1024000,
          "embedding_length": 5120
        },
        "capabilities": [
          "completion",
          "tools"
        ]
      },
      {
        "name": "glm-ocr:latest",
        "model": "glm-ocr:latest",
        "modified_at": "2026-05-24T00:39:52.652124252+02:00",
        "size": 2219299168,
        "digest": "6effedd0dc8aec66ad86c94b8099395376e212f22c87a9d7147c422f7e447be4",
        "details": {
          "parent_model": "",
          "format": "gguf",
          "family": "glmocr",
          "families": [
            "glmocr"
          ],
          "parameter_size": "1.1B",
          "quantization_level": "F16",
          "context_length": 131072,
          "embedding_length": 1536
        },
        "capabilities": [
          "vision",
          "completion",
          "tools"
        ]
      },
      {
        "name": "qwen2.5vl:7b",
        "model": "qwen2.5vl:7b",
        "modified_at": "2026-04-30T16:06:16.860402568+02:00",
        "size": 5969245856,
        "digest": "5ced39dfa4bac325dc183dd1e4febaa1c46b3ea28bce48896c8e69c1e79611cc",
        "details": {
          "parent_model": "",
          "format": "gguf",
          "family": "qwen25vl",
          "families": [
            "qwen25vl"
          ],
          "parameter_size": "8.3B",
          "quantization_level": "Q4_K_M",
          "context_length": 128000,
          "embedding_length": 3584
        },
        "capabilities": [
          "vision",
          "completion"
        ]
      },
      {
        "name": "hermes3:8b",
        "model": "hermes3:8b",
        "modified_at": "2026-04-29T15:16:30.945303841+02:00",
        "size": 4661227271,
        "digest": "5cebc002d642ae9045516c5fa7536434a568fcae89b79a9f7a60aab3bb8b3b33",
        "details": {
          "parent_model": "/var/lib/ollama/blobs/sha256-c8985d236593f7a17da2a3da49588aa951a9b1e57ce97753364fbf59e63af84a",
          "format": "gguf",
          "family": "llama",
          "families": [
            "llama"
          ],
          "parameter_size": "8.0B",
          "quantization_level": "Q4_0",
          "context_length": 131072,
          "embedding_length": 4096
        },
        "capabilities": [
          "completion",
          "tools"
        ]
      },
      {
        "name": "hermes4:14b",
        "model": "hermes4:14b",
        "modified_at": "2026-04-29T15:16:17.372464333+02:00",
        "size": 9001755604,
        "digest": "1d0bf06e40d5726c8bf2af9c986148e9ba7f703b94e54bba094012184c0e5977",
        "details": {
          "parent_model": "/var/lib/ollama/blobs/sha256-7ad9be1e446e3da0c149fdf55284c90be666d3e13c6e2581587853f4f9538073",
          "format": "gguf",
          "family": "qwen3",
          "families": [
            "qwen3"
          ],
          "parameter_size": "14.8B",
          "quantization_level": "Q4_K_M",
          "context_length": 40960,
          "embedding_length": 5120
        },
        "capabilities": [
          "completion",
          "tools",
          "thinking"
        ]
      },
      {
        "name": "qwen3:8b",
        "model": "qwen3:8b",
        "modified_at": "2026-04-29T15:16:14.936266201+02:00",
        "size": 5225388192,
        "digest": "a3d07590175f2ff95a0bb4f1828e8345dbe0f3606e62f4d9975f84f64b359922",
        "details": {
          "parent_model": "/var/lib/ollama/blobs/sha256-a3de86cd1c132c822487ededd47a324c50491393e6565cd14bafa40d0b8e686f",
          "format": "gguf",
          "family": "qwen3",
          "families": [
            "qwen3"
          ],
          "parameter_size": "8.2B",
          "quantization_level": "Q4_K_M",
          "context_length": 40960,
          "embedding_length": 4096
        },
        "capabilities": [
          "completion",
          "tools",
          "thinking"
        ]
      },
      {
        "name": "nomic-embed-text:latest",
        "model": "nomic-embed-text:latest",
        "modified_at": "2026-04-28T19:13:09.78155352+02:00",
        "size": 274302450,
        "digest": "0a109f422b47e3a30ba2b10eca18548e944e8a23073ee3f3e947efcf3c45e59f",
        "details": {
          "parent_model": "",
          "format": "gguf",
          "family": "nomic-bert",
          "families": [
            "nomic-bert"
          ],
          "parameter_size": "137M",
          "quantization_level": "F16",
          "context_length": 2048,
          "embedding_length": 768
        },
        "capabilities": [
          "embedding"
        ]
      }
    ]
  },
  "before": {
    "models": [
      {
        "name": "mistral-nemo:12b",
        "model": "mistral-nemo:12b",
        "size": 8246704209,
        "digest": "e7e06d107c6c86ed0cf45445f1790720b5092149c4c95f4d965844e9afbfdc89",
        "details": {
          "parent_model": "",
          "format": "gguf",
          "family": "llama",
          "families": [
            "llama"
          ],
          "parameter_size": "12.2B",
          "quantization_level": "Q4_0"
        },
        "expires_at": "2026-09-13T19:48:59.83465082+02:00",
        "size_vram": 8246704209,
        "context_length": 16384
      }
    ]
  },
  "options": {
    "num_ctx": 4096,
    "num_predict": 256,
    "temperature": 0,
    "seed": 42,
    "num_gpu": 99
  },
  "source": "https://docs.ollama.com/api/generate",
  "method": "one warmup + five repetitions per model, unique nonce prevents full prompt cache reuse, single serial request; cache effects retained in raw responses"
}