Benchmarks updated 5/5 sources ingestedLiveBench 2026-10-14 · Terminal-Bench 4.0 · OpenRouter prices · Artificial Analysis49 models · 35 Terminal-Bench runs

All models

Every model LiveBench has tested in its latest release. Sort by LiveBench overall or by price.

Showing 49 of 49 benchmark rows. Sorted by mean across axes — indicative only, scores are per-benchmark.

ModelOpenRouter idLiveBench overallLong AgenticReasoningCodingBrowserData AnalysisResearchDesign$/Mtok inContext
claude-fable-5.1effort not statedanthropic/claude-fable-5.188.080.891.797.7100.080.382.083.4$101000k
claude-opus-5.5effort not statedanthropic/claude-opus-5.586.585.092.296.380.080.385.086.9$41000k
gpt-6-astraeffort not statedopenai/gpt-6-astra86.182.492.793.784.083.085.282.0$101050k
gpt-6.1-soleffort not statedopenai/gpt-6.1-sol85.473.092.696.088.082.783.582.1$21050k
gemini-4-argoneffort not statedgoogle/gemini-4-argon84.169.995.189.792.085.283.173.7$21049k
claude-fable-5effort not statedanthropic/claude-fable-583.653.289.798.392.080.586.984.7$101000k
claude-sonnet-5.5effort not statedanthropic/claude-sonnet-5.582.578.891.694.784.059.583.685.0$21000k
claude-opus-5effort not statedanthropic/claude-opus-582.162.191.290.884.074.688.983.0$51000k
gpt-6-soleffort not statedopenai/gpt-6-sol81.966.988.793.192.081.269.382.2$21050k
deepseek-v4.1-flash-1017effort not stateddeepseek/deepseek-v4.1-flash-101779.250.989.196.076.079.579.583.4$0.31049k
muse-wonder-1.0effort not statedunmatched78.153.290.686.272.081.688.075.1$4—
deepseek-v4.1-flasheffort not stateddeepseek/deepseek-v4.1-flash78.040.186.790.284.079.386.879.2$0.171049k
claude-opus-4.8effort not statedanthropic/claude-opus-4.876.848.389.277.396.066.081.878.8$51000k
gpt-5.6-soleffort not statedopenai/gpt-5.6-sol76.348.591.789.788.079.854.581.6$41050k
claude-haiku-5.5effort not statedanthropic/claude-haiku-5.575.657.682.491.480.053.181.983.1$0.11000k
gpt-5.6-terraeffort not statedopenai/gpt-5.6-terra75.048.490.686.284.079.360.476.0$21050k
grok-4.6effort not statedx-ai/grok-4.674.940.791.289.164.075.384.579.8$2500k
mimo-v2.6-proeffort not statedxiaomi/mimo-v2.6-pro74.947.579.681.588.072.376.778.5$0.431050k
muse-spark-1.3effort not statedmeta/muse-spark-1.374.840.489.790.280.079.668.674.8$1.251049k
grok-4.7effort not statedx-ai/grok-4.774.150.182.784.560.076.986.478.0$2500k
minimax-m3.1-flasheffort not statedminimax/minimax-m3.1-flash73.134.782.983.992.071.472.774.0$0.151049k
claude-sonnet-5effort not statedanthropic/claude-sonnet-573.039.088.774.772.071.782.882.0$21000k
glm-5.3-flasheffort not statedz-ai/glm-5.3-flash70.628.477.680.684.076.470.177.2$0.011049k
grok-4.5effort not statedx-ai/grok-4.570.127.987.294.344.073.085.978.1$2500k
gpt-6-lunaeffort not statedopenai/gpt-6-luna69.448.181.884.572.073.443.582.5$0.11050k
qwen3.8-flasheffort not statedqwen/qwen3.8-flash68.837.187.487.448.074.274.672.8$0.151000k
gpt-5.5effort not statedopenai/gpt-5.568.744.689.785.148.081.661.670.3$51050k
gemini-3.8-flasheffort not statedgoogle/gemini-3.8-flash67.330.189.382.264.054.079.172.7$0.821049k
gemini-3.7-flasheffort not statedgoogle/gemini-3.7-flash67.228.387.880.556.068.076.573.2$0.751049k
gpt-5.6-lunaeffort not statedopenai/gpt-5.6-luna66.130.485.673.668.078.051.076.0$0.21050k
claude-sonnet-4.6effort not statedanthropic/claude-sonnet-4.665.123.186.466.668.076.163.671.8$31000k
qwen3.8-maxeffort not statedunmatched64.242.388.260.332.078.469.978.6$2—
deepseek-v4-flash-visioneffort not statedunmatched63.639.085.437.664.079.569.270.7$0.44—
kimi-k3effort not statedmoonshotai/kimi-k363.442.890.759.316.078.779.077.2$0.241049k
qwen3.8-27beffort not statedqwen/qwen3.8-27b62.727.980.086.832.076.659.475.9$0.21000k
muse-spark-1.2effort not statedmeta/muse-spark-1.262.421.390.078.236.076.567.767.1$1.251049k
step-5-previeweffort not statedstepfun/step-5-preview62.027.583.885.624.063.572.876.9$11049k
glm-5.3effort not statedz-ai/glm-5.360.735.985.882.80.070.273.277.3$0.081049k
muse-spark-openeffort not statedmeta/muse-spark-open57.618.973.773.636.060.280.360.3$1.251049k
beam-1.0effort not statedunmatched55.517.380.283.90.066.472.867.9$0.7—
gemini-3.1-proeffort not statedunmatched54.923.984.066.74.078.565.761.6$2—
minimax-m3effort not statedminimax/minimax-m353.712.474.570.98.076.268.066.2$0.31049k
inklingeffort not statedthinkingmachines/inkling51.914.178.364.016.072.859.958.3$1524k
nemotron-super-vl-3.5effort not statedunmatched51.515.878.865.54.064.670.960.7$0.23—
mistral-large-4effort not statedmistralai/mistral-large-4-047.711.883.960.08.076.528.165.6$1.361049k
claude-haiku-4.5effort not statedanthropic/claude-haiku-4.546.011.161.748.140.059.342.259.9$1200k
gemma-4-31beffort not statedgoogle/gemma-4-31b-it41.513.368.349.04.053.344.458.4$0.2262k
nemotron-ultra-3effort not statedunmatched40.811.274.743.60.054.543.258.6$0.6—
nemotron-lightning-3.5effort not statedunmatched28.78.342.922.60.043.140.043.9$0.06—