nordan/ai > Which Swedish Party Do LLMs Vote For?<br>Svenska Dagbladet recently put the big AI chatbots through SVT's Valkompass and reported which parties ChatGPT, Gemini, Claude and Grok picked. It is a nice idea. It also measures something narrower than the headline suggests.<br>The chat apps they used are not raw models. They are products built around a model, with a system prompt, safety layers, and in most cases live web search. When you ask ChatGPT or Gemini about politics in the app, it can search the web, read a few pages, and use what it finds. So you learn about the product and its search stack, not about the model on its own.<br>The chat window is also not where models do most of their work. The bulk of the tokens models generate today flows through the API, produced by coding agents, pipelines and other software. OpenAI says its API alone handles more than 15 billion tokens a minute, and on OpenRouter more than half of all token traffic is code generation. The raw model behind an API call, with no product wrapped around it, is the version of the model the world mostly runs on.<br>We were curious about the narrower question. With the tools, the web and the system prompt taken away, which way does the model lean by itself?<br>What we changed<br>We took the 35 Riksdag questions from SVT's Valkompass 2026 and ran every configuration on the Agent Arena leaderboard: 28 entries covering 23 frontier models from Anthropic, OpenAI, Google, xAI, DeepSeek, Moonshot, Z.ai, MiniMax, Alibaba and NVIDIA. Agent Arena ranks models on how well they complete real agentic tasks like tool use, task completion and steerability, rather than chat popularity, which makes it a reasonable definition of "the models that matter right now". Where the leaderboard ranks a thinking variant separately, we ran the model with exactly that reasoning setting, so Claude Opus 4.8 and Claude Opus 4.8 (Thinking) are separate rows here just as they are there. Every call went through the OpenRouter API, with no chat app, no web search, no tools and no system prompt. Only the weights.<br>For each configuration we compared its 35 answers to every party's official answers, then picked the party it sits closest to. The charts below have the full picture. (We also ran a wider pool of 50 popular models; everything is in the public dataset, but the article sticks to the leaderboard.)<br>What the models pick<br>All 28 Agent Arena leaderboard configurations, thinking settings included, and the party each one lands closest to.
1Claude Fable 5 (High)AnthropicLiberalerna<br>2Claude Opus 4.8 (Thinking)AnthropicLiberalerna<br>3GPT 5.5 (xHigh)OpenAIVänsterpartiet<br>4Claude Opus 4.7AnthropicModeraterna<br>5Claude Opus 4.7 (Thinking)AnthropicModeraterna<br>6GPT 5.5 (High)OpenAIVänsterpartiet<br>7GLM 5.2 (Max)ZhipuVänsterpartiet<br>8GPT 5.4 (High)OpenAIVänsterpartiet<br>9Claude Opus 4.6AnthropicSocialdemokraterna<br>10GPT 5.5OpenAIVänsterpartiet<br>11Claude Opus 4.8AnthropicLiberalerna<br>12Claude Sonnet 4.6AnthropicVänsterpartiet<br>13GLM 5.1ZhipuMiljöpartiet<br>14Kimi K2.7 CodeMoonshotMiljöpartiet<br>15Gemini 3.1 Pro PreviewGoogleSocialdemokraterna<br>16Gemini 3.5 FlashGoogleSocialdemokraterna<br>17DeepSeek V4 FlashDeepSeekMiljöpartiet<br>18Kimi K2.6MoonshotMiljöpartiet<br>19Minimax M3MiniMaxCenterpartiet<br>20DeepSeek V4 ProDeepSeekSocialdemokraterna<br>21Qwen 3.6 PlusAlibabaLiberalerna<br>22Grok 4.3 (High)xAICenterpartiet<br>23Grok Build 0.1xAIModeraterna<br>24Gemini 3 FlashGoogleModeraterna<br>25Minimax M2.7MiniMaxVänsterpartiet<br>26Nemotron 3 UltraNVIDIAMiljöpartiet<br>27Gemma 4 31BGoogleMiljöpartiet<br>28Grok 4.3xAIModeraterna
Every configuration on the Agent Arena leaderboard, in leaderboard order, named exactly as ranked there. Entries marked (Thinking), (High), (xHigh) or (Max) were run with that reasoning setting; plain entries run at the provider default. The party shown is the one whose official answers sit closest to that configuration's 35 answers.
Which parties the models lean toward<br>How much the leaderboard agrees with each party, and which party each configuration lands closest to.
Average agreement with each party, all 28 configurations<br>Centerpartiet
69.0%
Socialdemokraterna
68.8%
Vänsterpartiet
68.8%
Miljöpartiet
68.5%
Liberalerna
68.3%
Moderaterna
67.2%
Kristdemokraterna
56.7%
Sverigedemokraterna
48.7%
How closely the configurations' 35 answers match each party's, averaged over the whole leaderboard (0–100% scale). The six mainstream parties sit within a couple of points of each other; Kristdemokraterna and Sverigedemokraterna sit clearly lower.
Closest match: how many configurations land nearest each party<br>Vänsterpartiet
Miljöpartiet
Moderaterna
Socialdemokraterna
Liberalerna
Centerpartiet
Kristdemokraterna
Sverigedemokraterna
For each configuration we take its single best-matching party. None land closest to Kristdemokraterna or Sverigedemokraterna.
By company<br>The same agreement numbers rolled up per company, across its leaderboard configurations.
Company
Anthropic<br>7 configurations<br>68.01 pick<br>72.91...