PulseAugur
中
实时 20:43:25
English(EN) 📊 Molmo 7B-D — the actual numbers GPQA: 24% MMLU-Pro: 37.1% Humanity's Last Exam: 5.1% Long Context Reasoning: 0% Measured independently, not self-reported → ht

LLM基准测试结果：Molmo、Qwen3、Hermes和Mistral性能揭晓

作者 PulseAugur 编辑部 · [7 个来源] ·

独立基准测试揭示了多个大型语言模型在不同任务上的表现。Molmo 7B-D在GPQA和MMLU-Pro上得分较低，而Qwen3 Omni 30B A3B和Hermes 4 - Llama-3.1 405B在推理任务上展现出显著更强的能力。Mistral Medium 3在长上下文推理方面也表现强劲。 AI

影响 提供了多个LLM在关键基准测试上的对比性能数据，有助于为特定任务选择模型。

排序理由 报告了多个LLM的独立基准测试结果。

在 Mastodon — fosstodon.org 阅读 →

AI 生成摘要 · Google Gemini · 来自 7 个来源。 我们如何撰写摘要 →

LLM基准测试结果：Molmo、Qwen3、Hermes和Mistral性能揭晓

本文如何被排名

Signal score
0 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Research
报告了多个LLM的独立基准测试结果。
Source corroboration
7 independent sources
Strong cross-source corroboration — multiple independent publishers covered this within the clustering window.
Topics
model release, other
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
48 days old
Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.
Coverage growth since scoring
+3 source(s) since last score
New sources have picked up this story since our last re-score. Score will update on the next scoring pass.

完整方法见我们的编辑标准。

报道来源 [7]

  1. Mastodon — fosstodon.org TIER_1 English(EN) · [email protected] ·

    📊 Molmo 7B-D — 实际数字 GPQA: 24% MMLU-Pro: 37.1% 人类终极考试: 5.1% 长上下文推理: 0% 独立测量，非自我报告 → ht

    📊 Molmo 7B-D — the actual numbers GPQA: 24% MMLU-Pro: 37.1% Humanity's Last Exam: 5.1% Long Context Reasoning: 0% Measured independently, not self-reported → https:// olud.ai/leaderboard.html # LLM # Benchmarks # OpenSource # AI

    链接 olud.ai/leaderboard.html
  2. Mastodon — mastodon.social TIER_1 English(EN) · opensourceaitech ·

    📊 Devstral Small (25年5月) — 实际数字 GPQA: 43.4% MMLU-Pro: 63.2% Humanity's Last Exam: 4% Long Context Reasoning: 32% 独立测量，非自测

    📊 Devstral Small (May '25) — the actual numbers GPQA: 43.4% MMLU-Pro: 63.2% Humanity's Last Exam: 4% Long Context Reasoning: 32% Measured independently, not self-reported → https:// olud.ai/leaderboard.html # LLM # Benchmarks # OpenSource # AI

    链接 olud.ai/leaderboard.html
  3. Mastodon — mastodon.social TIER_1 English(EN) · opensourceaitech ·

    📊 Qwen3.5 9B (推理) — 实际数字 GPQA: 80.6% 人类最后考试: 14.9% 长上下文推理: 70% IFBench: 66.7% ⚡ 84.7 tokens/sec 💰 98 intellig

    📊 Qwen3.5 9B (Reasoning) — the actual numbers GPQA: 80.6% Humanity's Last Exam: 14.9% Long Context Reasoning: 70% IFBench: 66.7% ⚡ 84.7 tokens/sec 💰 98 intelligence points per dollar Measured independently, not self-reported → https:// olud.ai/leaderboard.html # LLM # Benchmarks …

    链接 olud.ai/leaderboard.html
  4. Mastodon — mastodon.social TIER_1 English(EN) · opensourceaitech ·

    📊 GLM-4.5V (推理) — 实际分数 GPQA: 68.4% MMLU-Pro: 78.8% 人类终极考试: 6.3% 长上下文推理: 0% 💰 每美元 10 智能点

    📊 GLM-4.5V (Reasoning) — the actual numbers GPQA: 68.4% MMLU-Pro: 78.8% Humanity's Last Exam: 6.3% Long Context Reasoning: 0% 💰 10 intelligence points per dollar Measured independently, not self-reported → https:// olud.ai/leaderboard.html # LLM # Benchmarks # OpenSource # AI

    链接 olud.ai/leaderboard.html
  5. Mastodon — mastodon.social TIER_1 English(EN) · opensourceaitech ·

    📊 Qwen3 Omni 30B A3B (推理) — 实际数字 GPQA: 72.6% MMLU-Pro: 79.2% 人类最后考试: 7.5% 长上下文推理: 0% ⚡ 110 tokens/秒 💰 22.1

    📊 Qwen3 Omni 30B A3B (Reasoning) — the actual numbers GPQA: 72.6% MMLU-Pro: 79.2% Humanity's Last Exam: 7.5% Long Context Reasoning: 0% ⚡ 110 tokens/sec 💰 22.1 intelligence points per dollar Measured independently, not self-reported → https:// olud.ai/leaderboard.html # LLM # Ben…

    链接 olud.ai/leaderboard.html
  6. Mastodon — mastodon.social TIER_1 English(EN) · opensourceaitech ·

    📊 Hermes 4 - Llama-3.1 405B (推理) — 实际分数 GPQA: 72.7% MMLU-Pro: 82.9% 人类最后考试: 10.9% 长上下文推理: 22.7% ⚡ 34.9 tokens

    📊 Hermes 4 - Llama-3.1 405B (Reasoning) — the actual numbers GPQA: 72.7% MMLU-Pro: 82.9% Humanity's Last Exam: 10.9% Long Context Reasoning: 22.7% ⚡ 34.9 tokens/sec 💰 5.9 intelligence points per dollar Measured independently, not self-reported → https:// olud.ai/leaderboard.html …

    链接 olud.ai/leaderboard.html
  7. Mastodon — mastodon.social TIER_1 English(EN) · opensourceaitech ·

    📊 Mistral Medium 3 — 实际数据 GPQA: 57.8% MMLU-Pro: 76% Humanity's Last Exam: 4.1% Long Context Reasoning: 31.7% 💰 每美元 15.6 智能点

    📊 Mistral Medium 3 — the actual numbers GPQA: 57.8% MMLU-Pro: 76% Humanity's Last Exam: 4.1% Long Context Reasoning: 31.7% 💰 15.6 intelligence points per dollar Measured independently, not self-reported → https:// olud.ai/leaderboard.html # LLM # Benchmarks # OpenSource # AI

    链接 olud.ai/leaderboard.html

相关实体

相关话题