第一项始终是图钉的来源。总体可信度是各份资料对上方所用开始和结束时间支持程度的加权平均;资料每比最新的一份旧 180 天,权重减半。
Epoch[5] AI将GLM-5的发布日期列为2026年2月11日,Maxime Labonne在Hugging Face[3][6]上的文章也写道“2026年2月11日……Z.ai正式发布了GLM-5”;Hugging Face上的权重自当天06:15 UTC起上传。The Decoder[4]的报道日期为2月12日,Z.ai的发布说明列为2026-02-12,这是同一次发布按北京时间计算的日期。
Gives the Artificial Analysis Intelligence Index v4.0 score of 50 (first open-weight model there), API prices of $1.00 and $3.20 per million tokens, and the 28-34% Hong Kong share jump on the day.[1]
Adds the "Pony Alpha" stealth run on OpenRouter from 6 February, LMArena Text Arena score 1452 (#1 open model) and the unconfirmed Reuters report of Huawei Ascend training.[1]
Independent report: 744B parameters, Vending Bench 2 result (run by Andon Labs), SWE-bench Verified 77.8 against Claude Opus 4.5's 80.9, document generation to .docx/.pdf/.xlsx, support for Huawei Ascend, Moore Threads and Cambricon chips.[1]
Epoch AI lists GLM-5 as an open-weights model from Z.ai released 11 February 2026, ranking 18th of 181 tracked models at release.
有遗漏或错误吗?用你自己的话说明:能佐证此图钉的链接、不同的开始或结束日期及理由,或缺失、有误的信息。AI 会对照此图钉的来源进行核实,搜索更好的来源,并添加任何支持你说法的页面。图钉自身的来源仍然最重要。AI 也会查看图片:显示的是别的东西或显示效果差的图片会被移到后面或替换。
原标题: GLM-5 Released
Z.ai模型卡中发布的结果表(原样转载)[6].
| 基准 | GLM-5 | GLM-4.7 | DeepSeek-V3.2 | Kimi K2.5 | Claude Opus 4.5 | Gemini 3 Pro | GPT-5.2 (xhigh) |
|---|---|---|---|---|---|---|---|
| HLE | 30.5 | 24.8 | 25.1 | 31.5 | 28.4 | 37.2 | 35.4 |
| HLE (使用工具) | 50.4 | 42.8 | 40.8 | 51.8 | 43.4* | 45.8* | 45.5* |
| AIME 2026 I | 92.7 | 92.9 | 92.7 | 92.5 | 93.3 | 90.6 | - |
| HMMT Nov. 2025 | 96.9 | 93.5 | 90.2 | 91.1 | 91.7 | 93.0 | 97.1 |
| IMOAnswerBench | 82.5 | 82.0 | 78.3 | 81.8 | 78.5 | 83.3 | 86.3 |
| GPQA-Diamond | 86.0 | 85.7 | 82.4 | 87.6 | 87.0 | 91.9 | 92.4 |
| SWE-bench Verified | 77.8 | 73.8 | 73.1 | 76.8 | 80.9 | 76.2 | 80.0 |
| SWE-bench Multilingual | 73.3 | 66.7 | 70.2 | 73.0 | 77.5 | 65.0 | 72.0 |
| Terminal-Bench 2.0 (Terminus 2) | 56.2 / 60.7 † | 41.0 | 39.3 | 50.8 | 59.3 | 54.2 | 54.0 |
| Terminal-Bench 2.0 (Claude Code) | 56.2 / 61.1 † | 32.8 | 46.4 | - | 57.9 | - | - |
| CyberGym | 43.2 | 23.5 | 17.3 | 41.3 | 50.6 | 39.9 | - |
| BrowseComp | 62.0 | 52.0 | 51.4 | 60.6 | 37.0 | 37.8 | - |
| BrowseComp (含上下文管理) | 75.9 | 67.5 | 67.6 | 74.9 | 67.8 | 59.2 | 65.8 |
| BrowseComp-Zh | 72.7 | 66.6 | 65.0 | 62.3 | 62.4 | 66.8 | 76.1 |
| τ²-Bench | 89.7 | 87.4 | 85.3 | 80.2 | 91.6 | 90.7 | 85.5 |
| MCP-Atlas (Public Set) | 67.8 | 52.0 | 62.2 | 63.8 | 65.2 | 66.6 | 68.0 |
| Tool-Decathlon | 38.0 | 23.8 | 35.2 | 27.8 | 43.5 | 36.4 | 46.3 |
| Vending Bench 2 | $4,432.12 | $2,376.82 | $1,034.00 | $1,198.46 | $4,967.06 | $5,478.16 | $3,591.33 |
*表示完整测试集上的结果;†表示修正了含糊指令的已验证Terminal-Bench 2.0变体。Vending Bench 2由Andon Labs运行。