LLM-Benchmarks
Von Wikiprompt, der freien Prompt-Enzyklopadie
Live-Ranglisten fur Sprachmodelle und generative KI, taglich aus offenen Community-Bewertungen aggregiert. Jede Tabelle nennt ihre Quelle, und da Wikiprompt taglich einen Snapshot speichert, bleiben Verschiebungen in den Rankings dokumentiert.
Zuletzt aktualisiert: 2026-10-13 · Dies ist die LIVE-Rangliste; datierte Snapshots bewahren jede fruhere Platzierung. Verlauf →
Interaktiver Explorer
Nach Arena, Organisation oder Lizenz filtern, Modelle suchen, Tabelle sortieren und das Kosten-Leistungs-Diagramm lesen. Die Tabellen darunter bleiben die zitierbare Referenz.
Sie sehen
LMArena · Text-Arena
Community-Elo aus abgestimmten Direktduellen auf LMArena.
Preise und Kontext: OpenRouter (Namensabgleich)
Text-Arena
Zeilen oder Diagrammpunkte anklicken, um Modelle zu fixieren und nur diese zu vergleichen.
Kosten vs. Leistung
Ausgabepreis in USD pro 1M Tokens (Log-Skala, OpenRouter) vs. Community-Elo (LMArena). Oben links = bestes Preis-Leistungs-Verhaltnis.
Attraktivster QuadrantPareto-LinieSkala:
Kuratierte Teilmenge (Top-Modelle je Organisation). Modelle wahlen oder Alle zeigen ankreuzen.
Elo im Zeitverlauf (Top-Textmodelle)
| Rang ▲ | Modell | Organisation | Elo-Wert | Stimmen | Eingabe $/1M | Ausgabe $/1M | Kontext |
|---|---|---|---|---|---|---|---|
| 4 | claude-opus-5-max | anthropic | 1,507 | 30,569 | $5 | $25 | 1,000,000 |
| 5 | claude-opus-4-6-high | anthropic | 1,503 | 79,415 | $4 | $20 | 1,000,000 |
| 6 | claude-opus-5-high | anthropic | 1,503 | 62,934 | $5 | $25 | 1,000,000 |
| 7 | gemini-3.8-flash-high | 1,499 | 31,437 | $0.75 | $3.75 | 1,048,576 | |
| 8 | claude-opus-4-6 | anthropic | 1,498 | 84,095 | $4 | $20 | 1,000,000 |
| 10 | claude-fable-5-high | anthropic | 1,491 | 40,026 | $10 | $50 | 1,000,000 |
| 11 | claude-opus-4-7-high | anthropic | 1,490 | 65,192 | $4 | $20 | 1,000,000 |
| 13 | gemini-3.7-flash-high | 1,486 | 24,173 | $0.75 | $3.75 | 1,048,576 | |
| 14 | qwen3.8-max | alibaba | 1,483 | 25,378 | |||
| 15 | claude-opus-4-7 | anthropic | 1,483 | 66,315 | $4 | $20 | 1,000,000 |
| 17 | gemini-3.5-flash-high | 1,482 | 48,533 | $1.5 | $9 | 1,048,576 | |
| 18 | gemini-3.1-pro-preview | 1,481 | 124,675 | $2 | $12 | 1,048,576 | |
| 19 | gemini-3-pro | 1,479 | 42,078 | $2 | $12 | 131,072 | |
| 20 | gemini-3.6-flash-high | 1,479 | 36,172 | $0.75 | $3.75 | 1,048,576 | |
| 21 | muse-spark-1.1 | meta | 1,479 | 39,847 | $1.25 | $4.25 | 1,048,576 |
| 24 | kimi-k3-max | moonshot | 1,476 | 30,351 | $0.73 | $10 | 1,048,576 |
| 26 | gpt-5.5-high | openai | 1,472 | 69,437 | $5 | $30 | 1,050,000 |
| 29 | glm-5.3-flash | zai | 1,471 | 27,678 | $0.03 | $0.9 | 1,048,576 |
| 33 | ernie-5.1 | baidu | 1,468 | 39,536 | |||
| 36 | mimo-v2.5-pro | xiaomi | 1,465 | 73,197 | $0.44 | $0.87 | 1,050,000 |
| 48 | deepseek-v4-pro | deepseek | 1,451 | 57,657 | $0.96 | $1.91 | 1,048,576 |
| 50 | grok-4.20-beta-0309-reasoning | xai | 1,451 | 66,524 | $1.25 | $2.5 | 2,000,000 |
| 54 | dola-seed-2.0-pro | bytedance | 1,448 | 79,195 | |||
| 72 | inkling | thinky | 1,441 | 37,744 | $1 | $4.05 | 524,288 |
| 90 | minimax-m3 | minimax | 1,433 | 61,220 | $0.3 | $1.2 | 1,048,576 |
| 97 | mistral-large-3 | mistral | 1,428 | 81,759 | $0.5 | $1.5 | 262,144 |
| 101 | longcat-flash-chat-2602-exp | meituan | 1,427 | 29,306 | |||
| 131 | amazon-nova-experimental-chat-11-10 | amazon | 1,415 | 26,059 |
Preise werden per normalisiertem Modellnamen dem nachsten OpenRouter-Listing zugeordnet; Reasoning-Varianten teilen den Basispreis.
Fazit
- • Aktueller Text-Arena-Fuhrer: claude-opus-5-max (anthropic, Elo 1507)
- • Bestes Preis-Leistungs-Verhaltnis (Elo pro Dollar, unter $2/1M Output): glm-5.3-flash (Elo 1471, $0.90/1M)
- • Starkstes offenes Modell: kimi-k3-max (Kimi K3 license, #24)
- • Coding-Fuhrer (Aider Pass Rate): gpt-5 (high) (88%)
Inhalt
Text-Arena25 ModelleKomplette Rangliste →
Quellen und Methodik: LMArena · Community-Elo aus abgestimmten Direktduellen auf LMArena.
| Rang | Modell | Organisation | Elo-Wert |
|---|---|---|---|
| 4 | claude-opus-5-max | anthropic | 1,507 |
| 5 | claude-opus-4-6-high | anthropic | 1,503 |
| 6 | claude-opus-5-high | anthropic | 1,503 |
| 7 | gemini-3.8-flash-high | 1,499 | |
| 8 | claude-opus-4-6 | anthropic | 1,498 |
| 10 | claude-fable-5-high | anthropic | 1,491 |
| 11 | claude-opus-4-7-high | anthropic | 1,490 |
| 13 | gemini-3.7-flash-high | 1,486 | |
| 14 | qwen3.8-max | alibaba | 1,483 |
| 15 | claude-opus-4-7 | anthropic | 1,483 |
| 17 | gemini-3.5-flash-high | 1,482 | |
| 18 | gemini-3.1-pro-preview | 1,481 | |
| 19 | gemini-3-pro | 1,479 | |
| 20 | gemini-3.6-flash-high | 1,479 | |
| 21 | muse-spark-1.1 | meta | 1,479 |
| 22 | gemini-3.5-flash-medium | 1,477 | |
| 24 | kimi-k3-max | moonshot | 1,476 |
| 26 | gpt-5.5-high | openai | 1,472 |
| 29 | glm-5.3-flash | zai | 1,471 |
| 30 | qwen3.5-max-preview | alibaba | 1,470 |
| 31 | glm-5.2-max | zai | 1,470 |
| 32 | gpt-5.4-high | openai | 1,469 |
| 33 | ernie-5.1 | baidu | 1,468 |
| 34 | gpt-5.5 | openai | 1,467 |
| 35 | gemini-3-flash | 1,466 |
Webentwicklungs-Arena25 ModelleKomplette Rangliste →
Quellen und Methodik: LMArena · Community-Elo aus abgestimmten Direktduellen auf LMArena.
| Rang | Modell | Organisation | Elo-Wert |
|---|---|---|---|
| 1 | claude-opus-5.5-max | anthropic | 1,813 |
| 2 | gpt-6-astra-max | openai | 1,786 |
| 3 | claude-sonnet-5.5-xhigh | anthropic | 1,774 |
| 4 | gpt-6.1-sol-max | openai | 1,755 |
| 5 | claude-fable-5.1-max | anthropic | 1,744 |
| 6 | claude-sonnet-5.5-high | anthropic | 1,715 |
| 7 | claude-opus-5-max | anthropic | 1,691 |
| 8 | gpt-6-sol-max | openai | 1,688 |
| 9 | gemini-4-argon-high | 1,678 | |
| 10 | qwen3.8-max-0902 | alibaba | 1,674 |
| 11 | qwen3.8-max | alibaba | 1,672 |
| 12 | claude-opus-5-high | anthropic | 1,657 |
| 13 | muse-spark-1.3-max | meta | 1,657 |
| 14 | kimi-k3-max | moonshot | 1,654 |
| 15 | grok-4.7-xhigh | xai | 1,639 |
| 16 | mimo-v2.6-flash | xiaomi | 1,637 |
| 17 | qwen3.8-flash-next | alibaba | 1,633 |
| 18 | hy4-preview | tencent | 1,632 |
| 19 | mimo-v2.6-pro | xiaomi | 1,629 |
| 20 | muse-spark-1.3 (xHigh) | meta | 1,626 |
| 21 | claude-fable-5-high | anthropic | 1,625 |
| 22 | glm-5.3-max | zai | 1,622 |
| 23 | deepseek-v4.1-flash-max | deepseek | 1,619 |
| 24 | gpt-5.6-sol-xhigh (codex-harness) | openai | 1,618 |
| 25 | grok-4.6-high | xai | 1,617 |
Vision-Arena25 ModelleKomplette Rangliste →
Quellen und Methodik: LMArena · Community-Elo aus abgestimmten Direktduellen auf LMArena.
| Rang | Modell | Organisation | Elo-Wert |
|---|---|---|---|
| 1 | claude-fable-5-high | anthropic | 1,324 |
| 2 | claude-opus-5.5-high | anthropic | 1,321 |
| 3 | claude-opus-5-high | anthropic | 1,319 |
| 4 | claude-fable-5.1-max | anthropic | 1,318 |
| 5 | claude-opus-4-7 | anthropic | 1,316 |
| 6 | gemini-3.7-flash-high | 1,316 | |
| 7 | claude-opus-4-6-high | anthropic | 1,316 |
| 8 | gemini-3.8-flash-high | 1,314 | |
| 9 | qwen3.8-max | alibaba | 1,314 |
| 10 | claude-opus-4-6 | anthropic | 1,313 |
| 11 | claude-opus-4-7-high | anthropic | 1,313 |
| 12 | gemini-3.5-flash-high | 1,310 | |
| 13 | muse-spark-1.3-max | meta | 1,309 |
| 14 | gemini-3.5-flash-medium | 1,309 | |
| 15 | muse-spark | meta | 1,306 |
| 16 | gemini-3-pro | 1,305 | |
| 17 | muse-spark-1.2 (xHigh) | meta | 1,305 |
| 18 | gpt-5.4-high | openai | 1,303 |
| 19 | muse-spark-1.3 (xHigh) | meta | 1,301 |
| 20 | gemini-3.6-flash-high | 1,298 | |
| 21 | gpt-5.5 | openai | 1,297 |
| 22 | gemini-3.1-pro-preview | 1,296 | |
| 23 | glm-5.3-flash | zai | 1,296 |
| 24 | claude-opus-4-8-high | anthropic | 1,294 |
| 25 | muse-spark-1.1 | meta | 1,293 |
Text-zu-Bild-Arena25 ModelleKomplette Rangliste →
Quellen und Methodik: LMArena · Community-Elo aus abgestimmten Direktduellen auf LMArena.
| Rang | Modell | Organisation | Elo-Wert |
|---|---|---|---|
| 1 | gpt-image-2.5-sunburst | openai | 1,425 |
| 2 | gpt-image-2.5-flare | openai | 1,398 |
| 3 | gpt-image-2 (medium) | openai | 1,383 |
| 4 | grok-imagine-image-2.0 (canvas) | xai | 1,334 |
| 5 | mai-image-2.6 | microsoft-ai | 1,332 |
| 6 | gemini-nano-banana-2.1 | 1,328 | |
| 7 | reve-2.1 | reve | 1,302 |
| 8 | grok-imagine-image-2.0 (20260801) | xai | 1,294 |
| 9 | muse-image | meta | 1,274 |
| 10 | reve-2.0 | reve | 1,269 |
| 11 | gemini-3.1-flash-image (nano-banana-2) [web-search] | 1,262 | |
| 12 | seedream-5.0-pro | bytedance | 1,255 |
| 13 | qwen-image-3.0-pro | alibaba | 1,255 |
| 14 | mai-image-2.5 | microsoft-ai | 1,254 |
| 15 | gemini-3.1-flash-lite-image (nano-banana-2-lite) | 1,250 | |
| 16 | gemini-3-pro-image-2k (nano-banana-pro) | 1,248 | |
| 17 | gpt-image-1.5-high-fidelity | openai | 1,237 |
| 18 | gemini-3-pro-image-preview (nano-banana-pro) | 1,234 | |
| 19 | qwen-image-2.1 | alibaba | 1,223 |
| 20 | ideogram-4.0-quality | ideogram | 1,205 |
| 21 | qwen-image-2.0-pro-2026-06-22 | alibaba | 1,192 |
| 22 | uni-1.1-max | luma-ai | 1,189 |
| 23 | uni-1.1 | luma-ai | 1,184 |
| 24 | mai-image-2 | microsoft-ai | 1,184 |
| 25 | grok-imagine-image | xai | 1,170 |
Bildbearbeitungs-Arena25 ModelleKomplette Rangliste →
Quellen und Methodik: LMArena · Community-Elo aus abgestimmten Direktduellen auf LMArena.
| Rang | Modell | Organisation | Elo-Wert |
|---|---|---|---|
| 1 | gpt-image-2.5-sunburst | openai | 1,524 |
| 2 | gpt-image-2.5-flare | openai | 1,481 |
| 3 | gpt-image-2 (medium) | openai | 1,462 |
| 4 | grok-imagine-image-2.0 (canvas) | xai | 1,439 |
| 5 | mai-image-2.6 | microsoft-ai | 1,428 |
| 6 | gemini-nano-banana-2.1 | 1,428 | |
| 7 | grok-imagine-image-2.0 (20260801) | xai | 1,425 |
| 8 | muse-image | meta | 1,403 |
| 9 | mai-image-2.5 | microsoft-ai | 1,401 |
| 10 | seedream-5.0-pro | bytedance | 1,394 |
| 11 | grok-imagine-image-quality (20260519) | xai | 1,391 |
| 12 | gemini-3-pro-image-2k (nano-banana-pro) | 1,390 | |
| 13 | chatgpt-image-latest-high-fidelity (20251216) | openai | 1,389 |
| 14 | gemini-3.1-flash-image (nano-banana-2) [web-search] | 1,387 | |
| 15 | gemini-3-pro-image-preview (nano-banana-pro) | 1,386 | |
| 16 | reve-2.1 | reve | 1,374 |
| 17 | gpt-image-1.5-high-fidelity | openai | 1,370 |
| 18 | qwen-image-2.1 | alibaba | 1,369 |
| 19 | reve-2.0 | reve | 1,359 |
| 20 | ideogram-4.5 | ideogram | 1,347 |
| 21 | uni-1.1-max | luma-ai | 1,334 |
| 22 | grok-imagine-image | xai | 1,329 |
| 23 | uni-1.1 | luma-ai | 1,315 |
| 24 | gemini-3.1-flash-lite-image (nano-banana-2-lite) | 1,314 | |
| 25 | qwen-image-2.0-pro-2026-06-22 | alibaba | 1,303 |
Text-zu-Video-Arena25 ModelleKomplette Rangliste →
Quellen und Methodik: LMArena · Community-Elo aus abgestimmten Direktduellen auf LMArena.
| Rang | Modell | Organisation | Elo-Wert |
|---|---|---|---|
| 1 | gemini-omni-1.1-flash | 1,516 | |
| 2 | gemini-omni-flash | 1,513 | |
| 3 | flux-3-video | bfl | 1,493 |
| 4 | grok-imagine-video-1.5-agent | xai | 1,492 |
| 5 | dreamina-seedance-2.0-720p | bytedance | 1,479 |
| 6 | wan3.0 | alibaba | 1,476 |
| 7 | dreamina-seedance-2.5-720p | bytedance | 1,474 |
| 8 | minimax-h3 | minimax | 1,460 |
| 9 | muse-video | meta | 1,456 |
| 10 | happyhorse-1.0 | aorizon | 1,427 |
| 11 | sora-2-pro | openai | 1,368 |
| 12 | veo-3.1-audio | 1,364 | |
| 13 | veo-3.1-audio-1080p | 1,362 | |
| 14 | veo-3.1-fast-audio | 1,362 | |
| 15 | veo-3.1-fast-audio-1080p | 1,358 | |
| 16 | veo-3-fast-audio | 1,348 | |
| 17 | sora-2 | openai | 1,343 |
| 18 | grok-imagine-video-720p | xai | 1,342 |
| 19 | veo-3-audio | 1,340 | |
| 20 | wan2.7-t2v | wan | 1,337 |
| 21 | wan2.6-t2v | alibaba | 1,327 |
| 22 | seedance-v1.5-pro | bytedance | 1,256 |
| 23 | veo-3 | 1,254 | |
| 24 | veo-3-fast | 1,248 | |
| 25 | wan2.5-t2v-preview | alibaba | 1,245 |
Videobearbeitungs-Arena10 ModelleKomplette Rangliste →
Quellen und Methodik: LMArena · Community-Elo aus abgestimmten Direktduellen auf LMArena.
| Rang | Modell | Organisation | Elo-Wert |
|---|---|---|---|
| 1 | wan3.0 | alibaba | 1,414 |
| 2 | dreamina-seedance-2.5-720p | bytedance | 1,410 |
| 3 | minimax-h3 | minimax | 1,392 |
| 4 | gemini-omni-flash | 1,367 | |
| 5 | dreamina-seedance-2.0-720p | bytedance | 1,365 |
| 6 | happyhorse-1.0 | aorizon | 1,307 |
| 7 | grok-imagine-video | xai | 1,258 |
| 8 | kling-o3-pro | kling | 1,255 |
| 9 | kling-o1-pro | kling | 1,197 |
| 10 | runway-gen4-aleph | runway | 1,182 |
Such-Arena25 ModelleKomplette Rangliste →
Quellen und Methodik: LMArena · Community-Elo aus abgestimmten Direktduellen auf LMArena.
| Rang | Modell | Organisation | Elo-Wert |
|---|---|---|---|
| 1 | gpt-5.6-sol-xhigh | openai | 1,257 |
| 2 | claude-opus-4-6-search | anthropic | 1,253 |
| 3 | gpt-5.5-search | openai | 1,242 |
| 4 | claude-opus-4-7 | anthropic | 1,233 |
| 5 | claude-fable-5 | anthropic | 1,230 |
| 6 | ernie-5.1 | baidu | 1,227 |
| 7 | claude-sonnet-4-6-search | anthropic | 1,221 |
| 8 | grok-4.5 | xai | 1,213 |
| 9 | gemini-3.1-pro-grounding | 1,210 | |
| 10 | gemini-3-pro-grounding | 1,207 | |
| 11 | gpt-5.2-search | openai | 1,207 |
| 12 | claude-opus-4-8 | anthropic | 1,204 |
| 13 | grok-4.20-multi-agent-beta-0309 | xai | 1,204 |
| 14 | gpt-5.1-search | openai | 1,199 |
| 15 | gemini-3-flash-grounding | 1,198 | |
| 16 | gpt-5.4-search | openai | 1,197 |
| 17 | claude-sonnet-5-search | anthropic | 1,194 |
| 18 | grok-4.20-beta1 | xai | 1,189 |
| 19 | claude-opus-4-5-search | anthropic | 1,180 |
| 20 | gpt-5.2-search-non-reasoning | openai | 1,172 |
| 21 | grok-4-1-fast-search | xai | 1,171 |
| 22 | grok-4-fast-search | xai | 1,171 |
| 23 | grok-4.3 | xai | 1,165 |
| 24 | claude-sonnet-4-5-search | anthropic | 1,158 |
| 25 | claude-opus-4-1-search | anthropic | 1,148 |
Agenten-Arena25 ModelleKomplette Rangliste →
Quellen und Methodik: LMArena · Community-Elo aus abgestimmten Direktduellen auf LMArena.
| Rang | Modell | Organisation | Elo-Wert |
|---|---|---|---|
| 1 | Claude Opus 5.5 (Max) | anthropic | |
| 2 | GPT 6.1 Sol (Max) | openai | |
| 3 | GPT 6 Sol (Max) | openai | |
| 4 | Gemini 4 Argon (High) | ||
| 5 | GPT 6 Astra (Max) | openai | |
| 6 | Claude Fable 5.1 (Max) | anthropic | |
| 7 | Claude Sonnet 5.5 (Max) | anthropic | |
| 8 | Claude Fable 5 (High) | anthropic | |
| 9 | Claude Opus 5 (Max) | anthropic | |
| 10 | Claude Opus 4.8 (High) | anthropic | |
| 11 | GPT 5.6 Sol (xHigh) | openai | |
| 12 | Gemini 3.8 Flash (High) | ||
| 13 | GPT 5.5 (xHigh) | openai | |
| 14 | Claude Sonnet 5 (High) | anthropic | |
| 15 | Grok 4.7 (xHigh) | xai | |
| 16 | Deepseek V4.1 Flash (Max) | deepseek | |
| 17 | Muse Spark 1.3 (Max) | meta | |
| 18 | MiMo V2.6 Pro | xiaomi | |
| 19 | GLM 5.2 (Max) | zai | |
| 20 | Kimi K3 (Max) | moonshot | |
| 21 | DeepSeek V4 Pro (High) (0813) | deepseek | |
| 22 | GPT 6 Luna (Max) | openai | |
| 23 | GPT 5.4 (High) | openai | |
| 24 | Hy4 preview | tencent | |
| 25 | MiMo V2.6 Flash | xiaomi |
Dokumenten-Arena25 ModelleKomplette Rangliste →
Quellen und Methodik: LMArena · Community-Elo aus abgestimmten Direktduellen auf LMArena.
| Rang | Modell | Organisation | Elo-Wert |
|---|---|---|---|
| 1 | claude-opus-5-high | anthropic | 1,516 |
| 2 | claude-fable-5.1-max | anthropic | 1,513 |
| 3 | claude-opus-4-6-high | anthropic | 1,507 |
| 4 | claude-opus-4-6 | anthropic | 1,507 |
| 5 | claude-fable-5 | anthropic | 1,496 |
| 6 | claude-opus-4-7 | anthropic | 1,495 |
| 7 | claude-opus-4-7-high | anthropic | 1,495 |
| 8 | gpt-5.5 | openai | 1,486 |
| 9 | gpt-5.5-high | openai | 1,484 |
| 10 | gpt-5.6-sol-xhigh | openai | 1,483 |
| 11 | claude-sonnet-4-6 | anthropic | 1,482 |
| 12 | claude-opus-4-8-high | anthropic | 1,475 |
| 13 | gpt-5.6-terra-xhigh | openai | 1,472 |
| 14 | gpt-5.4 | openai | 1,471 |
| 15 | muse-spark-1.3-max | meta | 1,471 |
| 16 | gpt-6-astra-max | openai | 1,468 |
| 17 | claude-sonnet-5-high | anthropic | 1,466 |
| 18 | muse-spark-1.1 | meta | 1,465 |
| 19 | claude-opus-4-8 | anthropic | 1,464 |
| 20 | gemini-3.5-flash-medium | 1,463 | |
| 21 | claude-opus-4-5-20251101 | anthropic | 1,462 |
| 22 | gemini-3.5-flash-high | 1,461 | |
| 23 | gpt-5.6-luna-xhigh | openai | 1,457 |
| 24 | gemini-3.6-flash-high | 1,456 | |
| 25 | grok-4.6-high | xai | 1,452 |
Weitere Ranglisten
BenchLM
Metrik: aggregated display score · benchlm.ai
| Rang | Modell | Organisation | Wert |
|---|---|---|---|
| 1 | Claude Opus 5.5 | Anthropic | 85.4 |
| 2 | GPT-6 Astra | OpenAI | 83.3 |
| 3 | Claude Sonnet 5.5 | Anthropic | 82.5 |
| 4 | GPT-6.1 Sol | OpenAI | 81.6 |
| 5 | Claude Fable 5.1 | Anthropic | 81.2 |
| 6 | Gemini 4 Argon | 80.6 | |
| 7 | Claude Fable 5 | Anthropic | 78.7 |
| 8 | Claude Opus 5 | Anthropic | 78.7 |
| 9 | GPT-5.6 Sol | OpenAI | 77.1 |
| 10 | GPT-6 Sol | OpenAI | 74.3 |
| 11 | Muse Wonder Full | Meta | 73.8 |
| 12 | MiMo-V2.6-Pro | Xiaomi | 73.1 |
| 13 | GPT-5.5 Pro | OpenAI | 72.6 |
| 14 | Gemini 3.8 Flash | 72.4 | |
| 15 | GPT-5.6 Terra | OpenAI | 71.8 |
LiveBench
Metrik: average score (0-100) · livebench.ai
| Rang | Modell | Organisation | Wert |
|---|---|---|---|
| 1 | claude-fable-5-1-max-effort | 83.8 | |
| 2 | claude-fable-5-max-effort | 83.4 | |
| 3 | claude-opus-5-5-max-effort | 83.2 | |
| 4 | gpt-6-astra-max | 83.1 | |
| 5 | gpt-6.1-sol-max | 82.5 | |
| 6 | muse-spark-1.3-xhigh | 82.4 | |
| 7 | claude-opus-5-5-xhigh-effort | 82.1 | |
| 8 | muse-wonder-1.0-max | 81.9 | |
| 9 | gpt-6.1-sol-xhigh | 81.8 | |
| 10 | gpt-5.6-sol-max | 81.7 | |
| 11 | deepseek-v4.1-flash-max | 81.5 | |
| 12 | gpt-5.5-xhigh | 80.8 | |
| 13 | claude-opus-5-max-effort | 80.5 | |
| 14 | gemini-3.7-flash-high | 79.9 | |
| 15 | gpt-6-sol-max | 79.8 |
Aider polyglot (coding)
Metrik: pass rate 2 (%) · aider.chat/docs/leaderboards/
| Rang | Modell | Organisation | Wert |
|---|---|---|---|
| 1 | gpt-5 (high) | 88 | |
| 2 | gpt-5 (medium) | 86.7 | |
| 3 | o3-pro (high) | 84.9 | |
| 4 | gemini-2.5-pro-preview-06-05 (32k think) | 83.1 | |
| 5 | o3 (high) | 81.3 | |
| 6 | gpt-5 (low) | 81.3 | |
| 7 | grok-4 (high) | 79.6 | |
| 8 | gemini-2.5-pro-preview-06-05 (default think) | 79.1 | |
| 9 | o3 (high) + gpt-4.1 | 78.2 | |
| 10 | Gemini 2.5 Pro Preview 05-06 | 76.9 | |
| 11 | o3 | 76.9 | |
| 12 | DeepSeek-V3.2-Exp (Reasoner) | 74.2 | |
| 13 | Gemini 2.5 Pro Preview 03-25 | 72.9 | |
| 14 | o4-mini (high) | 72 | |
| 15 | claude-opus-4-20250514 (32k thinking) | 72 |
Artificial Analysis (Intelligence Index)
Metrik: AA Intelligence Index (0-100) · artificialanalysis.ai
| Rang | Modell | Organisation | Wert |
|---|---|---|---|
| 1 | Claude Opus 5.5 (Max, Default Fallback) | Anthropic | 57.6 |
| 2 | Claude Sonnet 5.5 (Max, Default Fallback) | Anthropic | 56 |
| 3 | Claude Opus 5.5 (Xhigh, Default Fallback) | Anthropic | 56 |
| 4 | Claude Opus 5.5 (High, Default Fallback) | Anthropic | 53.6 |
| 5 | Claude Fable 5.1 (Max, Default Fallback) | Anthropic | 53.4 |
| 6 | Claude Fable 5.1 (Xhigh, Default Fallback) | Anthropic | 53.2 |
| 7 | GPT-6 Astra (Max) | OpenAI | 52.7 |
| 8 | Gemini 4 Argon (High) | 52.6 | |
| 9 | GPT-6 Astra (Xhigh) | OpenAI | 52.4 |
| 10 | Claude Sonnet 5.5 (Xhigh, Default Fallback) | Anthropic | 51.9 |
| 11 | GPT-6.1 Sol (Max) | OpenAI | 51.8 |
| 12 | Claude Opus 5.5 (Medium, Default Fallback) | Anthropic | 51.2 |
| 13 | Claude Fable 5.1 (High, Default Fallback) | Anthropic | 51.2 |
| 14 | GPT-6.1 Sol (Xhigh) | OpenAI | 51 |
| 15 | GPT-6 Astra (High) | OpenAI | 50.9 |
Preise und Kontextfenster
Quellen und Methodik: OpenRouter
Preise und Kontext-Metadaten aus der offentlichen OpenRouter-API. Preise in USD pro 1M Tokens.
| Modell | Organisation | Kontext | Eingabe $/1M | Ausgabe $/1M |
|---|---|---|---|---|
| Auto Router (Beta) | openrouter | 2,000,000 | $-1,000,000 | $-1,000,000 |
| Pareto Code Router | openrouter | 2,000,000 | $-1,000,000 | $-1,000,000 |
| SpaceXAI: Grok 4.20 Multi-Agent | x-ai | 2,000,000 | $1.25 | $2.5 |
| SpaceXAI: Grok 4.20 | x-ai | 2,000,000 | $1.25 | $2.5 |
| Auto Router | openrouter | 2,000,000 | $-1,000,000 | $-1,000,000 |
| Meta: Llama 4 Scout | meta-llama | 1,310,720 | $0.18 | $0.59 |
| OpenAI: GPT-6.1 Sol Pro | openai | 1,050,000 | $2 | $10 |
| OpenAI: GPT-6.1 Sol Pro (batch) | openai | 1,050,000 | $1 | $5 |
| OpenAI: GPT-6.1 Sol | openai | 1,050,000 | $2 | $10 |
| OpenAI: GPT-6.1 Sol (batch) | openai | 1,050,000 | $1 | $5 |
| OpenAI: GPT-6 Luna Pro | openai | 1,050,000 | $0.1 | $0.5 |
| OpenAI: GPT-6 Luna Pro (batch) | openai | 1,050,000 | $0.05 | $0.25 |
| OpenAI: GPT-6 Luna | openai | 1,050,000 | $0.1 | $0.5 |
| OpenAI: GPT-6 Luna (batch) | openai | 1,050,000 | $0.05 | $0.25 |
| OpenAI: GPT-6 Sol Pro | openai | 1,050,000 | $2 | $10 |
| OpenAI: GPT-6 Sol Pro (batch) | openai | 1,050,000 | $1 | $5 |
| OpenAI: GPT-6 Sol | openai | 1,050,000 | $2 | $10 |
| OpenAI: GPT-6 Sol (batch) | openai | 1,050,000 | $1 | $5 |
| Xiaomi: MiMo-V2.6-Flash | xiaomi | 1,050,000 | $0.14 | $0.28 |
| Xiaomi: MiMo-V2.6-Pro | xiaomi | 1,050,000 | $0.44 | $0.87 |
Letzte Anderungen
Bewegungen seit dem letzten Snapshot:
- GPT 6.1 Sol (Max) ▲ 5 → 2
- GPT 6 Sol (Max) ▲ 6 → 3
- Gemini 4 Argon (High) ▲ 7 → 4
- GPT 6 Astra (Max) ▼ 2 → 5
- Claude Fable 5.1 (Max) ▼ 3 → 6
- Claude Sonnet 5.5 (Max) ▼ 4 → 7
- Claude Opus 4.8 (High) ▲ 11 → 10
- GPT 5.6 Sol (xHigh) ▲ 12 → 11
- Gemini 3.8 Flash (High) ▲ 22 → 12
- GPT 5.5 (xHigh) ▲ 14 → 13
- Claude Sonnet 5 (High) ▼ 13 → 14
- Grok 4.7 (xHigh) ▲ 21 → 15
Quellen und Methodik
Die Ranglisten werden aus offenen Community-Bewertungen aggregiert. Elo-Werte, Konfidenzintervalle und Stimmen stammen von LMArena (Community-Votings in direkten Modell-Duellen, als offenes Dataset veroffentlicht). Preise und Kontext-Metadaten stammen aus der offentlichen OpenRouter-API. Wikiprompt erstellt taglich einen Snapshot und bewahrt die Historie; eigene Evaluationen fuhren wir nicht durch.