This page snapshots the current context-window specification for every major frontier LLM as of June 2026, organized by maximum advertised context size.
Context Window Snapshot
| Model | Vendor | Max Context | Type |
|---|---|---|---|
| Llama 4.5 Scout | Meta | 10,000,000 tokens | Open-weight |
| Gemini 3.2 Pro | 2,000,000 tokens | Closed | |
| Gemini 3.2 Flash | 2,000,000 tokens | Closed | |
| Llama 4.5 Maverick | Meta | 1,000,000 tokens | Open-weight |
| Claude Opus 4.7 (1M variant) | Anthropic | 1,000,000 tokens | Closed |
| Claude Sonnet 4.6 (1M variant) | Anthropic | 1,000,000 tokens | Closed |
| DeepSeek V4.1 Flash | DeepSeek | 1,000,000 tokens | Open + closed |
| DeepSeek V4.1 Pro | DeepSeek | 1,000,000 tokens | Closed |
| Qwen 3.7 (flagship) | Alibaba | 1,000,000 tokens | Open + closed |
| Hunyuan Large 3 | Tencent | 512,000 tokens | Closed + partial open |
| GPT-5.6 / Pro | OpenAI | 256,000 tokens | Closed |
| Mistral Medium 3 | Mistral AI | 256,000 tokens | Closed + self-host |
| ERNIE 5.1 | Baidu | 256,000 tokens | Closed |
| Doubao Pro (June 2026) | ByteDance | 256,000 tokens | Closed |
| GLM-6 | Zhipu AI | 256,000 tokens | Open |
| Claude Opus 4.7 (standard) | Anthropic | 200,000 tokens | Closed |
| Claude Sonnet 4.6 (standard) | Anthropic | 200,000 tokens | Closed |
| Claude Mythos 5 | Anthropic | 200,000 tokens | Closed |
| Claude Haiku 4.5 | Anthropic | 200,000 tokens | Closed |
Key Takeaways
- Llama 4.5 Scout at 10M tokens leads the industry by an order of magnitude.
- Gemini 3.2 Pro at 2M leads among closed models with retrieval quality refined in the 3.2 release.
- 1M context is now the open-weight frontier baseline (DeepSeek V4.1, Qwen 3.7, Llama 4.5 Maverick).
- 256K is the modal context size for the Chinese consumer-anchored frontier (ERNIE, Doubao, GLM, Hunyuan).
- Maximum advertised context does not equal usable context; retrieval quality at the upper end varies materially by model.
Methodology
Specs from vendor disclosures as of June 2026. Maximum context is the advertised ceiling; effective context (where retrieval quality degrades meaningfully) is typically 50 to 80% of maximum. Updated monthly.
How Presenc AI Helps
Presenc AI tracks how long-context capability shapes brand-visibility behavior. Long-context models reward authoritative long-form content and penalize thin marketing pages at the synthesis step.