Best AI model for long context
As of October 19, 2026, the largest context window TensorFeed tracks is Llama 4 Scout from Meta, at 10M tokens, an open-weights model. Among models with a list price, the largest is 1.05M tokens (GPT-6 Astra, GPT-6.1 Sol, GPT-6 Sol and GPT-6 Luna). Of the 61 models with a window of 1M tokens or more, Claude Opus 5.5 scores highest on the TensorFeed Intelligence Index's 2026 benchmark set (97.1), so it is the pick here.
Ordered by TensorFeed Intelligence Index score as of October 19, 2026; prices as of October 19, 2026.
Models with 1M tokens or more
TFII on the 2026 set first, then the 2024 set (ranked separately, never compared), then models the index does not rank yet.
| Model | Provider | Context | Price in / out | TFII | Terminal-Bench 4.0 |
|---|---|---|---|---|---|
| Claude Opus 5.5 | Anthropic | 1M | $4.00 / $20.00 | 97.1 | 65.15% |
| Claude Sonnet 5.5 | Anthropic | 1M | $2.00 / $10.00 | 93.7 | 64.14% |
| Claude Fable 5.1 | Anthropic | 1M | $10.00 / $50.00 | 89.9 | 58.08% |
| Claude Opus 5 | Anthropic | 1M | $5.00 / $25.00 | 83.9 | 53.54% |
| GPT-6 Astra | OpenAI | 1.05M | $10.00 / $50.00 | 83.3 | 59.6% |
| Claude Fable 5 | Anthropic | 1M | $10.00 / $50.00 | 81.6 | 41.41% |
| GPT-5.6 Sol | OpenAI | 1M | $4.00 / $20.00 | 74.5 | 37.88% |
| GLM-5.3 | Z.ai (Zhipu AI) | 1M | $1.40 / $4.40 | 71.8 | 38.89% |
| Claude Opus 4.8 | Anthropic | 1M | $5.00 / $25.00 | 67.9 | 23.23% |
| Kimi K3 | Moonshot AI | 1M | $3.00 / $15.00 | 65.1 | 17.17% |
| GLM-5.3 Flash | Z.ai (Zhipu AI) | 1M | $0.15 / $0.50 | 62.8 | 37.37% |
| GPT-5.6 Terra | OpenAI | 1M | $2.00 / $12.00 | 60.9 | 22.73% |
| Claude Sonnet 5 | Anthropic | 1M | $2.00 / $10.00 | 59.2 | 10.61% |
| GPT-5.6 Luna | OpenAI | 1M | $0.20 / $1.20 | 53.2 | 11.62% |
| MiniMax M3 | MiniMax | 1M | $0.30 / $1.20 | 29.8 | 1.01% |
| Gemini 3.8 Flash | 1M | $0.75 / $3.75 | 79.0 2024 set | 19.7% | |
| Gemini 3.6 Flash | 1M | $0.75 / $3.75 | 78.1 2024 set | - | |
| Muse Spark 1.1 | Meta | 1M | $1.25 / $4.25 | 78.1 2024 set | - |
| Qwen3.8-Max | Alibaba | 1M | $2.00 / $6.00 | 77.4 2024 set | 34.34% |
| GLM-5.2 | Z.ai (Zhipu AI) | 1M | $1.40 / $4.40 | 75.6 2024 set | - |
| Qwen3.8 27B | Alibaba | 1M | $0.50 / $3.00 | 75.2 2024 set | - |
| Gemini 3.5 Flash-Lite | 1M | $0.30 / $2.50 | 73.3 2024 set | - | |
| Claude Opus 4.7 | Anthropic | 1M | $5.00 / $25.00 | 69.8 2024 set | - |
| GPT-5.5 | OpenAI | 1M | $5.00 / $30.00 | 69.8 2024 set | - |
| DeepSeek V4 Pro | DeepSeek | 1M | $1.32 / $3.96 | 69.1 2024 set | - |
| Claude Opus 4.6 | Anthropic | 1M | $5.00 / $25.00 | 67.8 2024 set | - |
| Claude Sonnet 4.6 | Anthropic | 1M | $3.00 / $15.00 | 64.6 2024 set | - |
| Gemini 2.5 Pro | 1M | $1.25 / $10.00 | 64.3 2024 set | - | |
| Llama 4 Scout | Meta | 10M | open weights | 61.8 2024 set | - |
| Llama 4 Maverick | Meta | 1M | open weights | 56.1 2024 set | - |
| GPT-6.1 Sol | OpenAI | 1.05M | $2.00 / $10.00 | - | 55.05% |
| GPT-6 Sol | OpenAI | 1.05M | $2.00 / $10.00 | - | 44.44% |
| GPT-6 Luna | OpenAI | 1.05M | $0.10 / $0.50 | - | 13.64% |
| Gemini 4 Argon | 1M | $2.00 / $10.00 | - | 57.58% | |
| Gemini 3.1 Flash-Lite | 1M | $0.25 / $1.50 | - | - | |
| DeepSeek V4.1 Flash | DeepSeek | 1M | $0.30 / $1.20 | - | 19.7% |
| Qwen3.8 2.4T-A95B | Alibaba | 1M | $2.00 / $6.00 | - | - |
| GLM-5.3-FlashX | Z.ai (Zhipu AI) | 1M | $0.37 / $1.25 | - | - |
| Ember-1 | Fireworks AI | 1M | $3.00 / $15.00 | - | 20.71% |
| Laguna S 2.1 | poolside | 1M | $0.10 / $0.20 | - | - |
| Hy4 preview | Tencent | 1M | $0.83 / $2.50 | - | 8.08% |
| MiMo-V2.6-Pro | Xiaomi | 1M | $0.43 / $0.87 | - | 31.31% |
| MiMo-V2.6-Flash | Xiaomi | 1M | $0.14 / $0.28 | - | 24.24% |
| Claude Mythos 5.1 | Anthropic | 1M | $10.00 / $50.00 | - | - |
| Claude Haiku 5.5 | Anthropic | 1M | $0.10 / $0.50 | - | 36.36% |
| Muse Wonder | Meta | 1M | $4.00 / $20.00 | - | 49.49% |
| Muse Spark Open | Meta | 1M | $1.25 / $4.25 | - | 6.57% |
| Muse Spark 1.3 | Meta | 1M | $1.25 / $4.25 | - | 10.61% |
| Muse Spark 1.3 Contributor | Meta | 1M | $0.10 / $0.20 | - | - |
| Muse Spark 1.2 | Meta | 1M | $1.25 / $4.25 | - | 6.06% |
| Muse Spark 1.2 Contributor | Meta | 1M | $0.10 / $0.20 | - | - |
| Mistral Large 4 | Mistral | 1M | $0.68 / $2.09 | - | 27.78% |
| Qwen3.8 Max Prime | Alibaba | 1M | $4.00 / $12.00 | - | - |
| Qwen3.7-Max | Alibaba | 1M | $2.50 / $7.50 | - | - |
| Qwen3.8-Omni-Flash | Alibaba | 1M | $0.15 / $0.47 | - | - |
| Qwen3.8-Flash | Alibaba | 1M | $0.15 / $0.47 | - | - |
| Grok 4.3 | xAI | 1M | $1.25 / $2.50 | - | - |
| LongCat-2.0 | Meituan | 1M | $0.75 / $2.95 | - | - |
| LongCat-2.5-Preview | Meituan | 1M | $0.30 / $1.20 | - | - |
| Fugu Ultra v2.0 | Sakana AI | 1M | $5.00 / $30.00 | - | - |
| Fugu Max | Sakana AI | 1M | $2.00 / $6.00 | - | - |
Prices are list prices per 1M input / output tokens. TFII is the headline TensorFeed Intelligence Index score; scores marked 2024 set come from older benchmarks and are not comparable with the rest.
How this is ranked
Long-context work needs two things: a window big enough to hold the input, and a model strong enough to use it. This page lists every model TensorFeed tracks with a window of 1M tokens or more, ordered by headline TensorFeed Intelligence Index score.
A context window is an advertised maximum. It does not measure how well a model recalls or reasons over a long input, and TensorFeed does not yet track a long-context recall benchmark, so test retrieval on your own documents before you rely on the full window.
Questions
What is the best AI model for long context?
As of October 19, 2026, the largest context window TensorFeed tracks is Llama 4 Scout from Meta, at 10M tokens, an open-weights model. Among models with a list price, the largest is 1.05M tokens (GPT-6 Astra, GPT-6.1 Sol, GPT-6 Sol and GPT-6 Luna). Of the 61 models with a window of 1M tokens or more, Claude Opus 5.5 scores highest on the TensorFeed Intelligence Index's 2026 benchmark set (97.1), so it is the pick here.
How does TensorFeed pick the best AI model for long context?
Long-context work needs two things: a window big enough to hold the input, and a model strong enough to use it. This page lists every model TensorFeed tracks with a window of 1M tokens or more, ordered by headline TensorFeed Intelligence Index score.
Which AI model has the largest context window?
Of the models TensorFeed tracks, Llama 4 Scout from Meta has the largest advertised window, 10,000,000 tokens (as of October 19, 2026). An advertised window is a maximum, not a measure of recall over long inputs.