Best LLMs for Long-Context Reasoning
The best LLM for reasoning over long inputs is GPT-5.2-Codex, scoring 75.7% on long-context reasoning — a different question from how large a window it accepts. GPT-5 (75.6%) and GPT-5.1 (75.0%) round out the top three.
- 1
gpt-5.2-codexReasoningToolsJSON+140.1 intel · $1.75/M · 31 t/s75.7%LCR - 2gpt-5ReasoningToolsJSON+134.7 intel · $1.25/M · 65 t/s75.6%LCR
- 3gpt-5.1ReasoningToolsJSON+136.9 intel · $1.25/M · 60 t/s75.0%LCR
- 4
kimi-k3ReasoningToolsJSON+157.1 intel · $3.00/M · 64 t/s74.7%LCR - 5gpt-5.5ReasoningToolsJSON+154.8 intel · $5.00/M · 80 t/s74.3%LCR
- 6gpt-5.6-lunaReasoningToolsJSON+151.2 intel · $0.100/M · 111 t/s74.0%LCR
- 7gpt-5.6-terraReasoningToolsJSON+155.0 intel · $1.00/M · 83 t/s74.0%LCR
- 8
minimax-m3ReasoningToolsJSON+144.4 intel · $0.300/M · 113 t/s74.0%LCR - 9gpt-5.4ReasoningToolsJSON+151.4 intel · $2.50/M · 90 t/s74.0%LCR
- 10gpt-5.3-codexReasoningToolsJSON+144.3 intel · $1.75/M · 61 t/s74.0%LCR
- 11gpt-5.6-solReasoningToolsJSON+158.9 intel · $5.00/M · 51 t/s73.7%LCR
- 12
mimo-v2.5-proReasoningToolsJSON42.2 intel · $0.435/M · 86 t/s73.3%LCR - 13
gemini-3.1-pro-previewReasoningToolsJSON+246.5 intel · $2.00/M · 107 t/s72.7%LCR - 14gpt-5.2ReasoningToolsJSON+142.2 intel · $1.75/M · 81 t/s72.7%LCR
- 15
glm-5.2ReasoningToolsJSON51.1 intel · $0.760/M · 143 t/s71.3%LCR - 16
claude-sonnet-5ReasoningToolsJSON+153.4 intel · $2.00/M · 79 t/s70.7%LCR - 17claude-opus-4.7ReasoningToolsJSON+153.5 intel · $5.00/M · 82 t/s70.3%LCR
- 18claude-opus-5ReasoningToolsJSON+160.7 intel · $5.00/M · 79 t/s70.0%LCR
- 19claude-fable-5ReasoningToolsJSON+159.9 intel · $10.00/M · 64 t/s70.0%LCR
- 20gemini-3.6-flashReasoningToolsJSON+250.1 intel · $1.50/M · 135 t/s69.7%LCR
- 21
qwen3.6-max-previewReasoningToolsJSON40.0 intel · $1.03/M · 21 t/s69.7%LCR - 22kimi-k2.6ReasoningToolsJSON+144.2 intel · $0.589/M · 141 t/s69.7%LCR
- 23qwen3.6-plusReasoningToolsJSON+139.6 intel · $0.325/M · 37 t/s69.7%LCR
- 24gemini-3.5-flashReasoningToolsJSON+250.2 intel · $1.50/M · 270 t/s69.3%LCR
- 25gpt-5.4-miniReasoningToolsJSON+140.0 intel · $0.750/M · 181 t/s69.3%LCR
How this is ranked
AI models ranked by LCR — reasoning accuracy over long inputs, not just the size of the window they accept. A large context window is a capacity claim; this is the measurement of whether the model can still reason over information buried deep inside it. Pair it with the longest-context ranking, which sorts on raw window size.
Frequently asked
Which LLM reasons best over long inputs?
The best LLM for reasoning over long inputs is GPT-5.2-Codex, scoring 75.7% on long-context reasoning — a different question from how large a window it accepts. GPT-5 (75.6%) and GPT-5.1 (75.0%) round out the top three.
Which AI model handles long documents best?
The best AI model for reasoning over long inputs is GPT-5.2-Codex, scoring 75.7% on long-context reasoning — a different question from how large a window it accepts. GPT-5 (75.6%) and GPT-5.1 (75.0%) round out the top three.
What's a good alternative to GPT-5.2-Codex?
GPT-5 (75.6%) is the closest alternative on this metric, followed by GPT-5.1 (75.0%). See the full ranking above for the tradeoffs.