DeepSWE v1.1 · Higher is better · Scale 0–100%
+12 more in the full tableDeepSWE v1.1 leaderboard
Gemini 4 Argon leads at 77.9%. Blue bars route on MiniRouter today.
Full tableOne benchmark at a time. No composite scores.
Top five per board. Blue routes on MiniRouter.
DeepSWE v1.1 · Higher is better · Scale 0–100%
+12 more in the full tableGemini 4 Argon leads at 77.9%. Blue bars route on MiniRouter today.
Full tableTerminal Bench 2.1 · Higher is better · Scale 0–100%
+6 more in the full tableDeepSeek V4.1 Flash leads at 90.6%. Blue bars route on MiniRouter today.
Full tablePick a benchmark. Each score links to its source.
17 models · publisher-reported
| # | Model | Score |
|---|---|---|
| 1 | Gemini 4 ArgonGoogle | 77.9%Source ↗ |
| 2 | DeepSeek V4.1 Flash 1017DeepSeek | 77.0%Source ↗ |
| 3 | DeepSeek V4.1 FlashDeepSeek | 74.2%Source ↗ |
| 4 | GPT-5.6 SolOpenAI | 72.7%Source ↗ |
| 5 | Particle-1Particle AI | 72.57%Source ↗ |
| 6 | Grok 4.7SpaceXAI | 71.0%Source ↗ |
| 7 | Step 5 PreviewStepFun | 67.7%Source ↗ |
| 8 | GLM-5.3Z.ai | 66.9%Source ↗ |
| 9 | Grok 4.6xAI | 65.9%Source ↗ |
| 10 | Gemini 3.7 FlashGoogle | 65.3%Source ↗ |
| 11 | Kimi K3Moonshot AI | 63.8%Source ↗ |
| 12 | GLM-5.3-FlashZ.ai | 63.4%Source ↗ |
| 13 | GPT-5.4 miniOpenAI | 62.1%Source ↗ |
| 14 | Mistral Large 4Mistral AI | 61.7%Source ↗ |
| 15 | MiniMax M3MiniMax | 61.2%Source ↗ |
| 16 | Qwen3.8-Flash-NextAlibaba | 58.7%Source ↗ |
| 17 | Qwen3.8-2.4T-A95BAlibaba | 56.6%Source ↗ |
Higher scores rank first. Test setups vary by publisher. Prices blend input and output 3:1. Missing results are excluded.
DeepSWE v1.1 against current blended list price, 3:1 input to output. Up and left is better.
Scroll to see the full chart →
11 model families · first-party APIs
Output tokens per second · Higher is better. Source: Artificial Analysis ↗ · 2026-11-15. Not MiniRouter route measurements. One variant per family, picked by intelligence score. Prices blend input and output 3:1.
Score per $1M tokens · updated Nov 12, 2026
| # | Model / price tier | Score / $ |
|---|---|---|
| 1 | GLM-5.3-FlashLaunch discount | 533.9$0.12 / 1M |
| 2 | DeepSeek V4.1 Flash 1017Off-peak | 293.3$0.26 / 1M |
| 3 | DeepSeek V4.1 FlashOff-peak | 282.7$0.26 / 1M |
| 4 | GLM-5.3-FlashList | 266.9$0.24 / 1M |
| 5 | DeepSeek V4.1 Flash 1017Peak · weekdays 01–04 / 06–10 UTC | 146.7$0.52 / 1M |
| 6 | DeepSeek V4.1 FlashPeak · weekdays 01–04 / 06–10 UTC | 141.3$0.52 / 1M |
| 7 | MiniMax M3Standard | 87.4$0.70 / 1M |
| 8 | GPT-5.4 miniStandard | 78.9$0.79 / 1M |
| 9 | Kimi K3Standard | 60.8$1.05 / 1M |
| 10 | Step 5 PreviewVercel provider rates · checked Sep 21 | 47.5$1.43 / 1M |
| 11 | Gemini 3.7 FlashIntroductory | 43.5$1.50 / 1M |
| 12 | Grok 4.7Publisher starting price | 23.7$3.00 / 1M |
| 13 | Qwen3.8-2.4T-A95BModel Studio API | 22.9$2.48 / 1M |
| 14 | Grok 4.6Standard | 22.0$3.00 / 1M |
| 15 | Gemini 3.7 FlashStandard | 21.8$3.00 / 1M |
| 16 | Gemini 4 ArgonIntroductory publisher price | 19.5$4.00 / 1M |
| 17 | GPT-5.6 SolCurrent list price | 6.5$11.25 / 1M |
Reported score per $1M tokens at each published price, blended 3:1 input to output. Directional only. Unpriced: Particle-1 72.57%, GLM-5.3 66.9%, Mistral Large 4 61.7%, Qwen3.8-Flash-Next 58.7