LiveCodeBench v5
Created by UC Berkeley, MIT and Cornell, 14 Jan 2025 · Higher is better · Website
- Other labs
- Reported in a later post or card
Results
| Model | Score | Source | Source check |
|---|---|---|---|
| Nemotron 3 Nano Omni 30B-A3B28 Apr 2026 · NVIDIA | 63.2% | Nemotron 3 Nano Omni: Efficient and Open Multimodal Intelligenceproblems from Jul 2024 to May 2025; single run | |
| Nemotron 3 Super 120B-A12B11 Mar 2026 · NVIDIA | 81.19% | nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-FP8 model card (launch-day revision)problems from Jul to Dec 2024Our capture | |
| Nemotron 3 Nano 30B-A3B15 Dec 2025 · NVIDIA | 72.1% | Quantization-Aware Distillation for NVFP4 Inference Accuracy RecoveryBF16 baseline of an NVFP4 quantization study; reported later | |
| Magistral Medium10 Jun 2025 · Mistral | 59.4% | MagistralSource image | CorroboratedThe same number is printed in 1 other lab document: 1 |
| Magistral Small10 Jun 2025 · Mistral | 55.84% | mistralai/Magistral-Small-2506 model cardOur capture | |
| Mistral Medium 37 May 2025 · Mistral | 29.1% | Magistralreported laterSource image | CheckedA person checked this number against its source on 20 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced. |
| Qwen3-235B-A22B29 Apr 2025 · Alibaba | 70.7% | Qwen3: Think Deeper, Act Fasterproblems from Oct 2024 to Feb 2025Source image | CheckedA person checked this number against its source on 19 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced. |
| Gemini 2.5 Flash17 Apr 2025 · Google | 63.5% | Developers can now start building with Gemini 2.5 Flash.thinking; single attempt; pass@1; problems from 10/1/2024 to 2/1/2025Source image | CorroboratedThe same number is printed in 1 other lab document: 1 |
| Gemini 2.5 Pro25 Mar 2025 · Google | 70.4% | Gemini 2.5: Our most intelligent AI modelsingle attempt; pass@1; problems from 10/1/2024 to 2/1/2025Source image | CheckedA person checked this number against its source on 29 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced. |
| Gemini 2.0 Flash-Lite5 Feb 2025 · Google | 28.9% | Gemini 2.0 is now available to everyonepublic preview; problems from 10/01/2024 to 02/01/2025Source image | CheckedA person checked this number against its source on 29 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced. |
| Gemini 2.0 Pro5 Feb 2025 · Google | 36.0% | Gemini 2.0 is now available to everyoneexperimental version; problems from 10/01/2024 to 02/01/2025Source image | CorroboratedThe same number is printed in 1 other lab document: 1 |
| Gemini 2.0 Flash11 Dec 2024 · Google | 34.5% | Gemini 2.0 is now available to everyonegenerally available version; problems from 10/01/2024 to 02/01/2025; reported laterSource image |
