// Constants for all currently available GroqCloud models. // Use these instead of raw strings to avoid typos and get IDE completion. // // Example: // import gloq // import gloq/models // // gloq.default_groq_request() // |> gloq.with_key(api_key) // |> gloq.with_model(models.llama_3_1_8b_instant) // |> gloq.with_context("Hello!") // |> gloq.build() // ── Llama 3.3 ────────────────────────────────────────────────────────────── /// Llama 3.3 70B — latest generation, 128k context. Recommended default. pub const llama_3_3_70b_versatile = "llama-3.3-70b-versatile" /// Llama 3.3 70B with speculative decoding for faster output. pub const llama_3_3_70b_specdec = "llama-3.3-70b-specdec" // ── Llama 3.1 ────────────────────────────────────────────────────────────── /// Llama 3.1 8B — fast, lightweight, 128k context. Good for high-throughput. pub const llama_3_1_8b_instant = "llama-3.1-8b-instant" /// Llama 3.1 70B — powerful, 128k context. pub const llama_3_1_70b_versatile = "llama-3.1-70b-versatile" // ── Llama 3.2 ────────────────────────────────────────────────────────────── /// Llama 3.2 1B — ultra-lightweight model for edge use cases. pub const llama_3_2_1b_preview = "llama-3.2-1b-preview" /// Llama 3.2 3B — small but capable. pub const llama_3_2_3b_preview = "llama-3.2-3b-preview" /// Llama 3.2 11B Vision — multimodal model supporting image inputs. pub const llama_3_2_11b_vision_preview = "llama-3.2-11b-vision-preview" /// Llama 3.2 90B Vision — large multimodal model. pub const llama_3_2_90b_vision_preview = "llama-3.2-90b-vision-preview" // ── Llama 3 (legacy) ─────────────────────────────────────────────────────── /// Llama 3 8B — legacy model, 8k context. Prefer llama_3_1_8b_instant. pub const llama3_8b_8192 = "llama3-8b-8192" /// Llama 3 70B — legacy model, 8k context. Prefer llama_3_1_70b_versatile. pub const llama3_70b_8192 = "llama3-70b-8192" // ── Mixtral ──────────────────────────────────────────────────────────────── /// Mixtral 8x7B MoE — 32k context. Strong at reasoning and code. pub const mixtral_8x7b_32768 = "mixtral-8x7b-32768" // ── Gemma ────────────────────────────────────────────────────────────────── /// Gemma 2 9B — Google's efficient instruction-tuned model. pub const gemma2_9b_it = "gemma2-9b-it" /// Gemma 7B — legacy Google model. Prefer gemma2_9b_it. pub const gemma_7b_it = "gemma-7b-it" // ── DeepSeek ─────────────────────────────────────────────────────────────── /// DeepSeek R1 distilled on Llama 70B — strong reasoning model. pub const deepseek_r1_distill_llama_70b = "deepseek-r1-distill-llama-70b" // ── Qwen ─────────────────────────────────────────────────────────────────── /// Qwen QwQ 32B — strong reasoning and long-context model. pub const qwen_qwq_32b = "qwen-qwq-32b"