Datasets:
License:
| {"comparison": "deepseek-v4-flash-0731_vs_mistral-medium-3-5", "date": "2026-08-11", "metric": "intelligence_aa_index", "values": {"deepseek-v4-flash-0731": 52, "mistral-medium-3-5": 30}, "winner": "deepseek-v4-flash-0731", "source": "Artificial Analysis Intelligence Index v4.1.1"} | |
| {"comparison": "deepseek-v4-flash-0731_vs_mistral-medium-3-5", "date": "2026-08-11", "metric": "price_input_usd_per_mtok", "values": {"deepseek-v4-flash-0731": 0.14, "mistral-medium-3-5": 1.16}, "winner": "deepseek-v4-flash-0731", "note": "~8x moins cher en entrée", "source": "tarifs officiels"} | |
| {"comparison": "deepseek-v4-flash-0731_vs_mistral-medium-3-5", "date": "2026-08-11", "metric": "context_tokens", "values": {"deepseek-v4-flash-0731": 1000000, "mistral-medium-3-5": 256000}, "winner": "deepseek-v4-flash-0731", "note": "4x plus de contexte", "source": "fiches officielles"} | |
| {"comparison": "deepseek-v4-flash-0731_vs_mistral-medium-3-5", "date": "2026-08-11", "metric": "speed_output_tok_per_s", "values": {"deepseek-v4-flash-0731": 131, "mistral-medium-3-5": 147}, "winner": "mistral-medium-3-5", "source": "Artificial Analysis"} | |
| {"comparison": "deepseek-v4-flash-0731_vs_mistral-large-3", "date": "2026-08-11", "metric": "intelligence_aa_index", "values": {"deepseek-v4-flash-0731": 52, "mistral-large-3": 16}, "winner": "deepseek-v4-flash-0731", "note": "index AA mesuré déc. 2025 pour Mistral", "source": "Artificial Analysis"} | |
| {"comparison": "deepseek-v4-flash-0731_vs_mistral-large-3", "date": "2026-08-11", "metric": "benchalign_v5", "values": {"deepseek-v4-flash-0731": null, "mistral-large-3": 49.4}, "winner": null, "note": "DeepSeek non mesuré sur BenchLM", "source": "BenchLM"} | |
| {"comparison": "deepseek-v4-flash-0731_vs_nvidia-nemotron-3-5-lightning", "date": "2026-08-12", "metric": "intelligence_aa_index", "values": {"deepseek-v4-flash-0731": 52, "nvidia-nemotron-3-5-lightning": 24}, "winner": "deepseek-v4-flash-0731", "source": "Artificial Analysis v4.1.1"} | |
| {"comparison": "deepseek-v4-flash-0731_vs_nvidia-nemotron-3-5-lightning", "date": "2026-08-12", "metric": "role", "values": {"deepseek-v4-flash-0731": "frontalier généraliste/agent", "nvidia-nemotron-3-5-lightning": "exécution bas coût (sous-agent)"}, "winner": null, "note": "Complémentaires : routage plan→frontalier, exécution→Nemotron local", "source": "NVIDIA blog"} | |
| {"comparison": "deepseek-v4-flash-0731_vs_mimo-v2-5", "date": "2026-08-12", "metric": "intelligence_aa_index", "values": {"deepseek-v4-flash-0731": 52, "mimo-v2-5": 38}, "winner": "deepseek-v4-flash-0731", "source": "Artificial Analysis v4.1.1"} | |
| {"comparison": "deepseek-v4-flash-0731_vs_mimo-v2-5", "date": "2026-08-12", "metric": "speed_output_tok_per_s", "values": {"deepseek-v4-flash-0731": 125, "mimo-v2-5": 80}, "winner": "deepseek-v4-flash-0731", "note": "56% plus rapide ; latence 1,43s vs 2,84s", "source": "Artificial Analysis"} | |
| {"comparison": "deepseek-v4-flash-0731_vs_mimo-v2-5", "date": "2026-08-12", "metric": "swe_bench_verified", "values": {"deepseek-v4-flash-0731": 79.0, "mimo-v2-5": 71.0}, "winner": "deepseek-v4-flash-0731", "source": "SWE-bench Verified"} | |
| {"comparison": "deepseek-v4-flash-0731_vs_mimo-v2-5", "date": "2026-08-12", "metric": "terminal_bench_2_1", "values": {"deepseek-v4-flash-0731": 82.7, "mimo-v2-5": 60.7}, "winner": "deepseek-v4-flash-0731", "source": "Terminal-Bench 2.1"} | |
| {"comparison": "deepseek-v4-flash-0731_vs_mimo-v2-5", "date": "2026-08-12", "metric": "multimodal", "values": {"deepseek-v4-flash-0731": false, "mimo-v2-5": "natif texte/image/vidéo/audio"}, "winner": "mimo-v2-5", "note": "Seul avantage décisif de MiMo", "source": "fiches officielles"} | |
| {"comparison": "deepseek-v4-flash-0731_vs_mimo-v2-5", "date": "2026-08-12", "metric": "benchlm_public_score", "values": {"deepseek-v4-flash-0731": null, "mimo-v2-5": 58.3}, "winner": null, "note": "DeepSeek non classé publiquement", "source": "BenchLM"} | |