NanoGPT Speedrun Frontier

NanoGPT Speedrun Frontier Login Start training All models Best validated result for each model Filter by model 1 Fable 5 2,726 81.7% closed claude-code · high @24H 3,010 8.7 d 2 Opus 5 2,920 53.6% closed claude-code · max @24H 3,045 2.9 d 3 Kimi K3 2,930 52.2% closed prime-agent · max @24H 3,125 3.6 d 4 Kimi K3 2,974 45.8% closed kimi-code · max @24H 3,135 5.1 d 5 Opus 4.8 3,018 39.4% closed claude-code · max @24H 3,180 3.0 d 6 GPT-5.6 Sol 3,042 35.9% closed codex · xhigh @24H 3,160 6.1 d 7 GPT-5.6 Sol Pro 3,058 33.6% closed codex · xhigh @24H 3,100 3.4 d 8 Sonnet 5 3,105 26.8% closed claude-code · max @24H 3,120 2.0 d 9 GPT-5.6 Luna 3,110 26.1% closed codex · xhigh @24H 3,170 1.9 d 10 Grok 4.5 3,120 24.6% closed grok-cli · xhigh @24H 3,160 2.7 d 11 Qwen3.8 Max 3,120 24.6% closed qwen-code · max @24H 3,225 1.9 d 12 GLM 5.2 3,150 20.3% closed pi · high @24H 3,200 1.8 d 13 DeepSeek V4 Pro 3,205 12.3% closed claude-code · max @24H 3,205 1.1 d 14 GPT-5.6 Terra 3,214 11.0% closed codex · xhigh @24H 3,214 1.1 d 15 Grok 4.6 3,220 10.1% closed grok-cli · xhigh 0.6 d 16 Muse Spark 1.2 3,230 8.7% closed muse-code · xhigh 0.6 d 17 Muse Spark 1.1 3,232 8.4% closed pi · max @24H 3,240 3.7 d 18 GPT-5.5 3,234 8.1% closed codex · xhigh @24H 3,234 1.1 d 19 Kimi K2.7 3,240 7.2% closed kimi-code · max @24H 3,240 1.6 d 20 GLM 5.3 — no record claude-code · xhigh Model Harness Traces Fable 5 claude-code · high 2,726 81.7% 3,010 800M 1.1M 811 3k 8.7 Opus 5 claude-code · max 2,920 53.6% 3,045 183M 690k 292 401 2.9 Kimi K3 prime-agent · max 2,930 52.2% 3,125 112M 2.2M — 488 3.6 Kimi K3 kimi-code · max 2,974 45.8% 3,135 682M 1.4M 713 4k 5.1 Opus 4.8 claude-code · max 3,018 39.4% 3,180 318M 2.3M 427 2k 3.0 GPT-5.6 Sol codex · xhigh 3,042 35.9% 3,160 2.9B 2.2M 963 28k 6.1 GPT-5.6 Sol Pro codex · xhigh 3,058 33.6% 3,100 1.2B 4.6M 509 7k 3.4 Sonnet 5 claude-code · max 3,105 26.8% 3,120 998M 2.1M 213 2k 2.0 GPT-5.6 Luna codex · xhigh 3,110 26.1% 3,170 894M 888k 362 12k 1.9 Grok 4.5 grok-cli · xhigh 3,120 24.6% 3,160 46M 385k 399 4k 2.7 Qwen3.8 Max qwen-code · max 3,120 24.6% 3,225 216M 629k 312 866 1.9 GLM 5.2 pi · high 3,150 20.3% 3,200 57M 1.7M 194 1k 1.8 DeepSeek V4 Pro claude-code · max 3,205 12.3% 3,205 26M 319k 189 309 1.1 GPT-5.6 Terra codex · xhigh 3,214 11.0% 3,214 417M 298k 154 3k 1.1 Grok 4.6 grok-cli · xhigh 3,220 10.1% — 27M 346k 97 691 0.6 Muse Spark 1.2 muse-code · xhigh 3,230 8.7% — 41M 910k 56 724 0.6 — Muse Spark 1.1 pi · max 3,232 8.4% 3,240 122M 1.6M 489 2k 3.7 GPT-5.5 codex · xhigh 3,234 8.1% 3,234 70M 77k 185 614 1.1 Kimi K2.7 kimi-code · max 3,240 7.2% 3,240 160M 763k 187 3k 1.6 GLM 5.3 claude-code · xhigh — — — — — — — — — 6h Budget of 24 hours 9d Model Record Human 2,600 Baseline 3,290 3,010 3,045 3,100 3,120 3,125 3,135 3,160 3,160 3,170 3,180 3,200 3,205 3,214 3,220 3,225 3,230 3,234 3,240 3,240 — Gray runs ended before the selected budget Loading available trajectories… Open Traces to explore 41 curated full agent trajectories, including tool calls, subagents, and scratchpads. Be your own Lab. Start training Book a call Platform Lab Compute Research Company Careers 24 Merch Contact Community X LinkedIn Discord Luma Resources Docs Writings Events Merch Platform Status Terms Terms of Service Privacy Policy Security Policy © 2026 Prime Intellect, Inc.