Reference notes.
Info
Last reviewed: November 2026. Model versions, context windows, and pricing change rapidly — check provider docs for current specs.
Foundation models are large AI models trained on broad data that can be adapted to many downstream tasks. The term encompasses LLMs, vision models, and multimodal systems.
Commercial Models
Anthropic Claude
| Model | Context | Strengths |
|---|---|---|
| Claude Fable 5.2 | 1M | Flagship for open-ended technical reasoning, coding, and scientific research (Nov 2026) |
| Claude Mythos 5.2 | 1M | Same intelligence as Fable 5.2 with dual-use research access; gated behind verification programmes (Nov 2026) |
| Claude Opus 5.5 | 1M | Daily-driver flagship for heavy coding and long-running agent tasks |
| Claude Sonnet 5.5 | 1M | Best balance of speed, judgement, and capability |
| Claude Haiku 5.5 | 1M | Fastest, cost-effective tier |
- Adaptive and extended thinking for complex reasoning
- Up to 128K max output tokens
- Strong instruction following and tool use
- Constitutional AI training
- API Documentation
OpenAI GPT
| Model | Context | Strengths |
|---|---|---|
| GPT-6.2 Astra | ~1.05M | Flagship for complex reasoning, agentic workflows, scientific research, and coding (Nov 2026) |
| GPT-6.2 Sol | ~1.05M | High-volume, cost-efficient model for everyday coding, reviews, and professional workflows (Nov 2026) |
- Unified architecture with adjustable reasoning effort, superseding standalone
o3and GPT-5.x models - Up to 128K max output tokens
- Strong function calling, coding, and tool use
- API Documentation
Google Gemini
| Model | Context | Strengths |
|---|---|---|
| Gemini 4 Carbon | 1M+ | Deepest reasoning tier for long-horizon planning, complex software engineering, and science |
| Gemini 4 Neon | 1M+ | Mid-tier balance of deep reasoning and token efficiency (Nov 2026) |
| Gemini 4 Argon | 1M+ | General-purpose all-rounder for research and complex knowledge work |
| Gemini 4 Hydrogen | 1M+ | High-speed, cost-efficient model for everyday execution and high-volume agents |
- Native multimodal (text, images, audio, video)
- Configurable deep reasoning across tiers
- Strong agentic, coding, and scientific performance
- API Documentation
Others
- xAI Grok 4.7 — 500K context, adjustable reasoning effort, strong coding and long-horizon agentic execution
- Meta Muse — Meta’s proprietary frontier family (
Wonder,Spark), succeeding the open Llama line for frontier tasks (accompanied by the open-weight 30BGlimmervariant) - Qwen4-Max (Alibaba, Nov 2026) — Proprietary API-only flagship (2.4T total / 71B active MoE, 1M context, native multimodal); see Qwen 3.8 below for open-weight releases
- Cohere Command A — 111B parameters, 256K context, enterprise focus and RAG-optimised
- Amazon Nova 2 — AWS Bedrock native multimodal family (
Lite,Pro,Omni,Sonic), up to 1M context
Open-Source Models
Meta Llama
| Model | Parameters | Context | Notes |
|---|---|---|---|
| Llama 4 Scout | 109B total / 17B active (16 routed + 1 shared) | 10M | Largest context window available, iRoPE (April 2025) |
| Llama 4 Maverick | 400B total / 17B active (128 routed + 1 shared) | 1M | Larger expert pool (April 2025) |
| Llama 3.3 70B | 70B dense | 128K | Text-only instruct |
- Mixture-of-experts architecture (Llama 4); planned Llama 4 Behemoth was shelved as Meta shifted frontier development to its proprietary Muse family (
Wonder,Spark, with open-weight 30BGlimmer) - Native multimodal (text + images)
- Llama 4 Scout uses Interleaved RoPE (iRoPE) — trained at 256K, extrapolates to 10M
- Gated open licence (Llama Community Licence — not OSI open-source)
- Llama Downloads
Mistral
| Model | Parameters | Context | Notes |
|---|---|---|---|
| Mistral Large 4 | ~1.05T total / ~52B active (MoE) | 1M | Flagship open-weight sparse MoE, native vision, adjustable reasoning |
| Mistral Small 4 | 119B total / 6.5B active (MoE) | 256K | Compact hybrid reasoning and instruction MoE |
| Ministral 3 (3B/8B/14B) | Dense | 128K | Small models with vision baked in (base, instruct, reasoning) |
| Mistral OCR 4.1 | — | — | Document digitisation with layout structure and confidence scores |
- European AI company
- Strong efficiency/performance ratio
- Multilingual with native vision support
- Le Chat consumer product
- Mistral AI
Qwen (Alibaba)
| Model | Parameters | Context | Notes |
|---|---|---|---|
| Qwen 3.8 (27B / Omni-Flash) | 27B dense & multimodal Flash | 256K+ | Latest open-weight Qwen generation |
| Qwen3-235B-A22B | 235B total / 22B active (MoE) | 128K | Widely deployed open MoE |
| Qwen3-VL | Various | 128K+ | Vision-language family |
| QwQ-32B | 32B | 128K | Earlier reasoning baseline |
- Flagship Qwen4-Max (Nov 2026; 2.4T total / 71B active MoE, 1M context, native multimodal) is API-only; Qwen 3.8 and earlier generations remain open-weight under Apache 2.0
- Strong multilingual (100+ languages)
- Extensive model family (code, maths, embedding, reranking variants)
- Qwen3-Embedding series leads open-weight multilingual retrieval on MTEB
- Qwen
DeepSeek
| Model | Parameters | Notes |
|---|---|---|
| DeepSeek-V4-Pro | ~1.6T MoE (49B active) | Native 1M context, integrated reasoning |
| DeepSeek-V4-Flash / V4.1-Flash | 284B (13B active) / 552B multimodal MoE | Fast, cost-efficient variants with native 1M context |
| DeepSeek-V3.2 / R1 | 671B MoE (37B active) | Earlier open-weight baselines with distilled variants |
- Competitive with frontier models at a fraction of training and inference cost
- Cost-efficient MoE architecture with multi-head latent attention and integrated reasoning
- Open weights under MIT licence — free for commercial use
- DeepSeek
Others
- Gemma 4 (Google, April 2026) — Open weights under Apache 2.0; spans
E2B,E4B,12B(multimodal including audio),26BMoE, and31Bdense - OLMo 3 / 3.1 (AI2) — Fully open (weights, training code, checkpoints, and Dolma 3 data) in 7B and 32B (
Base,Instruct,Think) under Apache 2.0 - Phi-4 (Microsoft) — Small dense model with strong reasoning
- GPT-OSS 20B / 120B (OpenAI, Aug 2025) — OpenAI’s open-weight reasoning release under Apache 2.0;
gpt-oss-120btargets a single 80 GB data-centre GPU (e.g. H100), whereasgpt-oss-20bis the consumer/on-device variant
Model Comparison Factors
Capability Benchmarks
See Evaluation & Benchmarking for details.
- Humanity’s Last Exam (HLE) — Frontier cross-disciplinary reasoning and expert knowledge
- GPQA Diamond — Graduate-level science
- SWE-bench Pro — Complex real-world software engineering
- ARC-AGI-2 / ARC-AGI-3 — Novel abstract reasoning and generalisation
- Arena — Crowdsourced human preference rankings (replacing saturated legacy benchmarks like MMLU, HumanEval, and GSM8K)
Practical Considerations
| Factor | Considerations |
|---|---|
| Latency | Time to first token, tokens/second |
| Cost | Per-token pricing, volume discounts |
| Context | How much text can be processed |
| Reliability | Uptime, consistency |
| Privacy | Data handling, compliance |
| Ecosystem | SDKs, documentation, support |
Licence Types
- Proprietary API — No access to weights (GPT-6.2, Claude 5.5 / Fable 5.2, Gemini 4, Qwen4-Max, Grok 4.7)
- Gated open — Weights available with restrictions (Llama 4)
- Permissive open — Few restrictions (Mistral Large 4 / Small 4, Qwen 3.8, DeepSeek-V4, Gemma 4, GPT-OSS)
- Fully open — Weights, code, and training data (OLMo 3 / 3.1)
API Providers
Model Providers
Direct from the source:
Aggregators / Routers
Access multiple models through one API:
- OpenRouter — Multiple providers, unified API
- Together AI — Open models, fine-tuning
- Fireworks AI — Fast inference
- Groq — Ultra-fast inference
Cloud Platforms
- AWS Bedrock — Multiple providers in AWS
- Azure OpenAI — OpenAI in Azure
- Google Vertex AI — Gemini and others
See Model Serving for self-hosted inference and deployment options.
Choosing a Model
Decision Framework
- Task requirements — What capability is most important?
- Latency needs — Real-time vs batch processing
- Cost constraints — Budget per million tokens
- Privacy requirements — Can data leave your environment?
- Context needs — How much text per request?
- Compliance — Regulatory requirements
Rules of Thumb
- Start with a capable model (Claude Sonnet 5.5 / Opus 5.5, GPT-6.2 Astra / Sol, Gemini 4 Argon / Neon)
- Optimise for cost/speed once it works (Haiku 5.5, GPT-6.2 Sol, Gemini 4 Hydrogen, DeepSeek-V4-Flash)
- Open models for privacy-sensitive use cases (DeepSeek-V4, Mistral Large 4 / Small 4, Qwen 3.8, Gemma 4, Llama 4)
- Frontier API pricing continues to drop rapidly per unit of capability — re-evaluate cost assumptions regularly
- Smaller models for high-volume, simple tasks
Staying Current
The landscape changes rapidly. Track developments:
- Artificial Analysis — Performance and cost benchmarks
- Arena — Human preference rankings (formerly LMSYS Chatbot Arena)
- Hugging Face Open LLM Leaderboard — Open-weight evaluations (v1 archived)
- Model provider blogs and announcements
Resources
- Foundation Models Paper — Stanford’s influential survey
- State of GPT — Karpathy’s overview
- Arena — Live human-preference leaderboard
- Open LLM Leaderboard