News-Tracker

Modeller

AI & LLM · daglig briefing


Ændringer siden forrige snapshot

Ny i indekset

  • Qwen3 32B (Reasoning) er ny i Artificial Analysis' indeks (intelligence 8.6, Alibaba).
  • Qwen3 14B (Reasoning) er ny i Artificial Analysis' indeks (intelligence 8.2, Alibaba).
  • Qwen3 8B (Reasoning) er ny i Artificial Analysis' indeks (intelligence 7.3, Alibaba).

Score ændret

  • Ling 3.0 Flash: intelligence 20.6 → 24.9 (+4.3).
  • Gemma 4 31B (Reasoning): intelligence 15.4 → 19 (+3.6).
  • Ling 3.0 Tiny: intelligence 11.9 → 15.3 (+3.4).
  • MiMo-V2.5: intelligence 22.3 → 25.2 (+2.9).
  • Mercury 2: intelligence 11.5 → 13.8 (+2.3).
  • Inkling Small: intelligence 26.1 → 27.8 (+1.7).
  • Llama 4 Scout: intelligence 6.5 → 8.1 (+1.6).
  • DeepSeek V4 Flash 0420 (Reasoning, High Effort): intelligence 24.8 → 26 (+1.2).
  • Mistral Small 3.2: intelligence 7 → 8.2 (+1.2).
  • Agnes 2.5 Pro Alpha: intelligence 27.8 → 26.8 (-1.0).
  • Qwen3 Coder Next: intelligence 10.1 → 9.2 (-0.9).
  • Devstral Small 2: intelligence 8.4 → 7.5 (-0.9).
  • GPT-5.5 Instant (June 2026): intelligence 26.8 → 26 (-0.8).
  • Command A+: intelligence 13.9 → 13.1 (-0.8).
  • Nemotron 3 Super 120B A12B (Reasoning): intelligence 13.6 → 12.8 (-0.8).
  • Devstral 2: intelligence 9.4 → 8.6 (-0.8).
  • Gemini 3.1 Pro Preview: intelligence 30.4 → 29.7 (-0.7).
  • Claude 4.5 Haiku (Reasoning): intelligence 17.6 → 16.9 (-0.7).
  • Ring-2.6-1T: intelligence 17.3 → 16.6 (-0.7).
  • Mistral Medium 3.5: intelligence 14.9 → 14.2 (-0.7).
  • Nemotron 3.5 Lightning: intelligence 13.6 → 12.9 (-0.7).
  • gpt-oss-120b (high): intelligence 12.3 → 11.6 (-0.7).
  • Granite 4.2 8B: intelligence 11.8 → 11.1 (-0.7).
  • Llama 4 Maverick: intelligence 9.3 → 10 (+0.7).
  • Muse Spark 1.1 (xhigh): intelligence 34.3 → 33.7 (-0.6).
  • Qwen3.7 Plus: intelligence 25.8 → 25.2 (-0.6).
  • LongCat 2.0: intelligence 19.7 → 19.1 (-0.6).
  • Qwen3.6 35B A3B (Reasoning): intelligence 18.8 → 18.2 (-0.6).
  • Muse Glimmer (high): intelligence 18.1 → 17.5 (-0.6).
  • Gemini 2.5 Pro: intelligence 16.7 → 16.1 (-0.6).
  • K2 Horizon 3.7B: intelligence 16.2 → 15.6 (-0.6).
  • DeepSeek V3.1 Terminus (Reasoning): intelligence 15.4 → 14.8 (-0.6).
  • MiniCPM5-2B: intelligence 13.1 → 12.5 (-0.6).
  • DeepSeek V4 Pro 0424 (Reasoning, Max Effort): intelligence 30.9 → 30.4 (-0.5).
  • Kimi K2.6: intelligence 27.5 → 27 (-0.5).
  • Kimi K2.7 Code: intelligence 26.3 → 25.8 (-0.5).
  • Hy3: intelligence 25.8 → 25.3 (-0.5).
  • Inkling (xhigh): intelligence 25.5 → 25 (-0.5).
  • Grok 4.3 (high): intelligence 25.4 → 24.9 (-0.5).
  • GPT-5.4 mini (xhigh): intelligence 24.6 → 24.1 (-0.5).
  • Nemotron 3 Ultra 550B A55B (Reasoning): intelligence 23.4 → 22.9 (-0.5).
  • Gemini 3.5 Flash-Lite: intelligence 22.7 → 22.2 (-0.5).
  • Qwen3.6 27B (Reasoning): intelligence 21.9 → 21.4 (-0.5).
  • Claude 4.5 Sonnet (Reasoning): intelligence 21.2 → 20.7 (-0.5).
  • GPT-5.4 nano (xhigh): intelligence 21.2 → 20.7 (-0.5).

Ny i HF-trending

  • XGENlabs/XGEN-JING er ny i Hugging Faces trending (score 66).
  • UntMods/FaceSwap_MiniMaxH3_REF2VA er ny i Hugging Faces trending (score 64).
  • TokenRhythm/NeoHorse-1-4B er ny i Hugging Faces trending (score 58.400000000000006).
  • cua-ai/cua-s1-forms er ny i Hugging Faces trending (score 56).

Intelligence Index — Artificial Analysis

#ModelIntel.Agentic CreatorUdgivet
1 Claude Fable 5.1 (Adaptive Reasoning, Max Effort, Default Fallback) 5 effort-varianter 53.4 57.9 Anthropic 2026-09-01
2 GPT-6 Astra (max) 5 effort-varianter 52.7 51 OpenAI 2026-09-03
3 Claude Opus 5 (Adaptive Reasoning, Max Effort) 5 effort-varianter 50.8 56.5 Anthropic 2026-07-24
4 Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback) 49.6 50.7 Anthropic 2026-06-09
5 Muse Spark 1.3 (max) 2 effort-varianter 48.1 55.5 Meta 2026-09-02
6 GPT-5.6 Sol (max) 6 effort-varianter 47 50.2 OpenAI 2026-07-09
7 Qwen3.8 Max (0902) 2 effort-varianter 45.4 56 Alibaba 2026-09-02
8 GLM-5.3 (max) ⬇ hf.co/zai-org/GLM-5.3 44.8 53.1 Z AI 2026-08-18
9 Grok 4.6 (high) 4 effort-varianter 44.3 53 SpaceXAI 2026-08-12
10 Step 5 Preview ⬇ hf.co/stepfun-ai/Step-5-Preview 43.7 StepFun 2026-09-18
11 Kimi K3 (max) 2 effort-varianter ⬇ hf.co/moonshotai/Kimi-K3 43.6 50 Kimi 2026-07-16
12 GPT-5.6 Terra (max) 6 effort-varianter 42.1 43.2 OpenAI 2026-07-09
13 Claude Opus 4.8 (Adaptive Reasoning, Max Effort) 41.8 41.9 Anthropic 2026-05-28
14 GLM 5.3 Flash ⬇ hf.co/zai-org/GLM-5.3-Flash 41.8 50.9 Z AI 2026-08-26
15 Gemini 3.8 Flash (high) 3 effort-varianter 40.9 40.2 Google 2026-09-02
16 Claude Opus 4.7 (Adaptive Reasoning, Max Effort) 2 effort-varianter 40.7 38.6 Anthropic 2026-04-16
17 Qwen3.8 2.4T A95B ⬇ hf.co/Qwen/Qwen3.8-2.4T-A95B 39.9 50.1 Alibaba 2026-08-12
18 Qwen3.8-Flash-Next ⬇ hf.co/Qwen/Qwen3.8-Flash-Next 39.8 53.6 Alibaba 2026-08-26
19 Gemini 3.7 Flash (medium) 3 effort-varianter 39.6 Google 2026-08-13
20 Muse Spark 1.2 (xhigh) 39.6 43.2 Meta 2026-08-05
21 DeepSeek V4.1 Flash (Reasoning, Max Effort) ⬇ hf.co/deepseek-ai/DeepSeek-V4.1-Flash 39.5 DeepSeek 2026-09-10
22 GPT-5.4 (xhigh) 3 effort-varianter 39 OpenAI 2026-03-05
23 Grok 4.5 (high) 38.8 41.2 SpaceXAI 2026-07-08
24 GPT-5.5 (xhigh) 5 effort-varianter 38.4 36.4 OpenAI 2026-04-23
25 Claude Sonnet 5 (Adaptive Reasoning, Max Effort) 6 effort-varianter 38.2 43.6 Anthropic 2026-06-30
26 GPT-5.6 Luna (max) 6 effort-varianter 37.3 42.1 OpenAI 2026-07-09
27 DeepSeek V4 Pro 0813 (Reasoning, Max Effort) ⬇ hf.co/deepseek-ai/DeepSeek-V4-Pro 36 41.3 DeepSeek 2026-08-13
28 Agnes 3.0 Flash ⬇ hf.co/Agnes-AI/Agnes-3.0-Flash 35.5 Sapiens AI 2026-09-11
29 Agnes 2.5 Pro Beta 35.2 Sapiens AI 2026-08-26
30 DeepSeek V4 Flash Vision (Reasoning, Max Effort) 34.8 47.5 DeepSeek 2026-08-21
31 DeepSeek V4 Flash 0731 (Reasoning, Max Effort) ⬇ hf.co/deepseek-ai/DeepSeek-V4-Flash-0731 34.3 41 DeepSeek 2026-07-31
32 Gemini 3.6 Flash (high) 34 29 Google 2026-07-21
33 GLM-5.2 (max) 2 effort-varianter ⬇ hf.co/zai-org/GLM-5.2 33.7 38.4 Z AI 2026-06-16
34 Muse Spark 1.1 (xhigh) 33.7 25.8 Meta 2026-07-09
35 Qwen3.8 27B (xhigh) 4 effort-varianter ⬇ hf.co/Qwen/Qwen3.8-27B 33.7 45.8 Alibaba 2026-08-14
36 Gemini 3.5 Flash (medium) 3 effort-varianter 33.6 Google 2026-05-19
37 Motif 3 2 effort-varianter ⬇ hf.co/Motif-Technologies/Motif-3 33.6 Motif Technologies 2026-08-12
38 GPT-5.3 Codex (xhigh) 32.5 OpenAI 2026-02-05
39 Claude Opus 4.6 (Adaptive Reasoning, Max Effort) 2 effort-varianter 31.9 Anthropic 2026-02-05
40 Muse Spark 31.3 Meta 2026-04-08

Hugging Face — trending familier

#Rod-repoTrendingLikes DownloadsParamsAlder
1 deepseek-ai/DeepSeek-V4.1-Flash +3 derivater 958 3324 482270 763.2B 10 d
2 Qwen/Qwen3.8-27B +13 derivater 619 15776 7365368 27.8B 46 d
3 XingChen-AGI/Xing4.0-29B-A4B +1 derivater 591 666 7278 31.2B 4 d
4 convaiinnovations/laya 583 591 0 421.3M 2 d
5 m-a-p/YuE2-3B +1 derivater 528 883 15446 3.6B 10 d
6 Lightricks/LTX-2.5 435 4455 1607815 59 d
7 ukisai/Swift-Qwen3.8-27b +1 derivater 435 493 8761 27.8B 11 d
8 harshatheg/Qwen-2.5-1B-RLCD 424 439 0 4 d
9 openbmb/MiniCPM5-2B +1 derivater 311 1600 389555 2.5B 14 d
10 Qwen/Qwen3.8-Flash-Next +2 derivater 259 5460 742586 180.0B 27 d
11 prism-ml/Ternary-Bonsai-2-27B-mlx-2bit 246 257 23111 27.4B 3 d
12 meta-llama/Llama-3.1-8B +1 derivater 234 7753 5919746 8.0B 794 d
13 AlexWortega/openjev 221 223 0 3 d
14 Qwen/Qwen3.6-35B-A3B +1 derivater 214 3496 68403 34.7B 11 d
15 MiniMaxAI/MiniMax-H3 +5 derivater 209 5494 4299737 33.1B 54 d

Params er safetensors.total, altså lagrede vægte inklusive kvantiserede tensorer — ikke et mål for hvad modellen kræver af hardware.