210 results
Foundation models you can call or deploy
by DeepSeek·Apr 2026
DeepSeek V4 preview: MoE with ~1M-token context via YaRN, FP8/FP4 quantization, MIT license; targets highly efficient million-token intelligence.
by Mistral AI·May 2024
Mistral's 22B code-specialized dense LLM under Mistral Non-Production License; supports 80+ programming languages, fill-in-the-middle and code completion.
by Anthropic·Mar 2024
Smallest, fastest model in Anthropic's Claude 3 family for high-volume tasks; supports vision input and the Messages API. Retired April 2026.
by xAI·Aug 2026
A coding and knowledge-work model designed for long-running agents, multi-step reasoning, and visual or interactive tasks.
by Mistral AI·Feb 2024
Mistral's small-tier LLM family; original API model launched with Mistral Large in 2024, later refreshed as the 24B Apache 2.0 open Mistral Small 3 and 3.1.
by Meta·Apr 2025
Meta's Llama 4 Scout MoE model with 17B active / 16 experts (109B total), 10M token context, natively multimodal, under the Llama 4 Community License.
by Microsoft·Feb 2025
Microsoft's compact 3.8B Phi-4 SLM designed for on-device inference; released alongside Phi-4-multimodal under an MIT license with 128K context support.
Meta's fourth-generation Llama family using MoE: Scout (17B active/16 experts) and Maverick (17B active/128 experts), natively multimodal, Llama 4 Community License.
by Anthropic·Oct 2025
Small, fast model in Anthropic's Claude 4.5 family for high-volume tasks; 200k-token context, supports extended thinking, priced $1/$5 per MTok.
by Z.ai·Jan 2026
30B/3B-active MoE (glm4_moe_lite) with 128K context, MIT license; small/fast GLM-4.7 tier for local and agentic coding with speculative decoding.
by Mistral AI·May 2025
Mistral's commercial mid-tier LLM served via API; modern line launched with Mistral Medium 3 targeting enterprise SOTA at ~8x lower cost, later refreshed as 3.1 and 3.5.
by xAI·Aug 2024
xAI's second-generation LLM, launched in beta on X with Grok-2 and Grok-2 mini; weights later open-sourced in 2025, using dense attention with MoE elements.
by Google·May 2026
Gemini 3.5 Flash: fast agentic and coding model built for long-horizon tasks, rivaling flagship-model quality at Flash-tier latency and cost.
by Midjourney·Apr 2025
Midjourney's V7 text-to-image model with improved aesthetics, prompt adherence, and image detail; accessible via Discord and the Midjourney web app.
by Google·Mar 2026
Fastest and cheapest Gemini 3 series model, aimed at high-volume workloads with adjustable thinking levels, 1M-token context, and multimodal input.
by Moonshot AI·Jan 2026
1T/32B-active MoE with 256K context and MoonViT vision, Modified MIT license; native multimodal with agent-swarm parallel sub-task orchestration.
Mistral's flagship closed-weights dense LLM (~123B in v2), 32k context; French-built rival to GPT-4/Claude, available via la Plateforme and Azure AI.
by Meta·Jul 2024
Meta's Llama 3.1 open-weight family with 8B, 70B, and 405B dense variants extended to a 128K token context, released under the Llama 3.1 Community License.
158B DeepSeek V4 Flash MoE with ~1M-token context, MIT license; smaller efficient tier of V4 designed for local and agentic inference workflows.
by Alibaba·Aug 2025
Text-to-image diffusion foundation with high-fidelity Chinese/English text rendering, Apache-2.0 license; supports style transfer, editing, and multi-aspect.
by Meta·Apr 2024
Meta's third-generation open-weight LLM family with 8B and 70B dense variants, 8K token context, base and Instruct, under the Llama 3 Community License.
Meta's 8B dense open-weight LLM from Llama 3.1, context extended to 128K tokens from Llama 3's 8K, released July 2024 under the Llama 3.1 Community License.
by OpenAI·Aug 2025
OpenAI's open-weight 120B-parameter reasoning model released under Apache 2.0, aimed at production self-hosting with function calling and structured outputs support.
by Mistral AI·Dec 2023
Mistral's sparse Mixture-of-Experts family; Mixtral 8x7B (46.7B total, ~12.9B active) under Apache 2.0, outperforming Llama 2 70B and GPT-3.5 on many benchmarks.
