Kostenlose Modelle mit langem Kontext
Modelle mit einem Kontextfenster von 256K Token oder mehr — für RAG ohne Chunking, Reviews ganzer Repositories und lange Dokumente.
65 Modelle11 Anbieterbis zu 10.5M Kontextbeste: GLM 5.3Aktualisiert:
Weiterlesen
Kostenlose Routen begrenzen den Kontext oft unter das Maximum des Modells. Die Spalte zeigt das anbieterspezifische Limit.
| # | Modell ↕ | AA Index ↕ | GPQA ↕ | Kontext ↕ | Größe ↕ | Kostenlos bei ↕ | |
|---|---|---|---|---|---|---|---|
| 1 |  |
8.1 | 59% | 10.5M | Sehr groß |  Together AI |
|
| 2 |  |
7.9 | 59% | 2.1M | Sehr groß | Google AI Studio |
|
| 3 | 19.1 | 78% | 1M | Sehr groß |  Nous Portal |
||
| 4 |  |
44.8 | 92% | 1M | Sehr groß |  Together AI, NVIDIA Build |
|
| 5 | |
33.7 | 90% | 1M | Sehr groß | Together AI |
|
| 6 | 27.8 | 90% | 1M | Klein |  OpenRouter |
||
| 7 |  |
22.9 | 87% | 1M | Sehr groß |  OpenRouter, NVIDIA Build, UnoRouter |
|
| 8 | 25.0 | 87% | 1M | Klein | OpenRouter |
||
| 9 | |
— | — | 1M | Sehr groß | Together AI |
|
| 10 |  |
11.7 | 70% | 1M | Sehr groß | Together AI |
|
| 11 | |
10.0 | 68% | 1M | Sehr groß | Together AI |
|
| 12 | — | — | 1M | Klein |  OpenRouter, Vercel AI Gateway, Nous Portal, UnoRouter |
||
| 13 | |
8.9 | 62% | 1M | Mittelgroß | Google AI Studio |
|
| 14 | |
7.3 | 54% | 1M | Klein | Google AI Studio |
|
| 15 | |
7.1 | 46% | 1M | Mittelgroß | Google AI Studio |
|
| 16 | 46.3 | — | 1M | Sehr groß | UnoRouter |
||
| 17 |  |
43.6 | 94% | 1M | Sehr groß | NVIDIA Build, UnoRouter |
|
| 18 | |
41.8 | 91% | 1M | Mittelgroß |  OrcaRouter, NVIDIA Build, UnoRouter |
|
| 19 |  |
39.8 | 92% | 1M | Mittelgroß | UnoRouter |
|
| 20 |  |
34.3 | 91% | 1M | Sehr groß | OrcaRouter, UnoRouter |
|
| 21 | |
34.0 | 93% | 1M | Mittelgroß | UnoRouter |
|
| 22 | |
22.2 | 84% | 1M | Klein | UnoRouter |
|
| 23 | |
12.9 | 74% | 1M | Mittelgroß | OpenRouter, NVIDIA Build, UnoRouter |
|
| 24 | |
— | — | 1M | Mittelgroß | UnoRouter |
|
| 25 |
Space Bunny Alpha
Stealth model · New
|
— | — | 1M | Klein | OpenRouter, UnoRouter |
|
| 26 |
Mimo V2.6 Flash
New
|
— | — | 1M | Klein | UnoRouter |
|
| 27 |  |
28.2 | 89% | 524K | Sehr groß | Nous Portal |
|
| 28 |
K2 Horizon
New
|
— | — | 524K | Klein | UnoRouter |
|
| 29 |
Axon 1.8 Pro
New
|
— | — | 512K | Sehr groß | UnoRouter |
|
| 30 |
Dots3-Note Preview
Dots Studio
|
— | — | 512K | Klein | OpenRouter, UnoRouter |
|
| 31 |
Axon 1.8 Flash
New
|
— | — | 512K | Klein | UnoRouter |
|
| 32 | — | — | 512K | Klein | UnoRouter |
||
| 33 | |
33.7 | 91% | 262K | Mittelgroß |  OpenRouter, Groq, UnoRouter |
|
| 34 | 22.6 | — | 262K | Klein | OpenRouter, Nous Portal, UnoRouter |
||
| 35 |  |
19.0 | 86% | 262K | Mittelgroß | OpenRouter, Together AI, NVIDIA Build, UnoRouter |
|
| 36 | |
16.7 | 79% | 262K | Mittelgroß | OpenRouter, Together AI, UnoRouter |
|
| 37 |  |
19.5 | 81% | 262K | Klein | Nous Portal |
|
| 38 | |
19.3 | 85% | 262K | Mittelgroß | Together AI |
|
| 39 | |
18.2 | 84% | 262K | Mittelgroß | Together AI |
|
| 40 | |
17.7 | 83% | 262K | Sehr groß | Together AI |
|
| 41 | |
12.8 | 80% | 262K | Sehr groß | OpenRouter, NVIDIA Build, UnoRouter |
|
| 42 | |
10.3 | 47% | 262K | Mittelgroß |  +1OpenRouter, Together AI, NVIDIA Build, TokenRouter, UnoRouter |
|
| 43 | |
14.2 | 75% | 262K | Mittelgroß | Together AI |
|
| 44 | |
13.7 | 81% | 262K | Klein | Together AI |
|
| 45 | |
13.4 | 77% | 262K | Sehr groß | Together AI |
|
| 46 | |
12.8 | 80% | 262K | Sehr groß | Together AI |
|
| 47 | |
12.7 | 79% | 262K | Sehr groß | Together AI |
|
| 48 |  |
9.9 | 76% | 262K | Klein | OpenRouter, UnoRouter |
|
| 49 | |
9.8 | 71% | 262K | Mittelgroß | Together AI |
|
| 50 | |
9.6 | 52% | 262K | Sehr groß | Together AI |
|
| 51 | — | — | 262K | Klein | OpenRouter, Vercel AI Gateway, Nous Portal, UnoRouter |
||
| 52 | — | — | 262K | Klein | OpenRouter, NVIDIA Build, Nous Portal, UnoRouter |
||
| 53 | |
9.5 | 67% | 262K | Mittelgroß | UnoRouter |
|
| 54 | — | — | 262K | Mittelgroß | Together AI |
||
| 55 |
Holo3 35B A3B
35B · 3B
|
— | — | 262K | Mittelgroß | Together AI |
|
| 56 | — | — | 262K | Mittelgroß | UnoRouter |
||
| 57 | |
— | — | 262K | Mittelgroß | UnoRouter |
|
| 58 | |
8.9 | 76% | 262K | Mittelgroß | Together AI |
|
| 59 | |
8.8 | 67% | 262K | Klein | Together AI |
|
| 60 | |
6.9 | 46% | 262K | Klein | Together AI |
|
| 61 | — | — | 262K | Klein | UnoRouter |
||
| 62 |  |
— | — | 256K | Sehr groß | UnoRouter |
|
| 63 | |
5.8 | 38% | 256K | Mittelgroß | Mistral AI, UnoRouter |
|
| 64 | |
— | — | 256K | Klein | Mistral AI |
|
| 65 | |
— | — | 256K | Klein | UnoRouter |
Keine Modelle entsprechen Ihren Filtern.
Benchmarks: Artificial Analysis (24. September 2026). Ein Strich bedeutet, dass das Modell nicht in deren Rangliste steht; wir schätzen Werte nie selbst.
FAQ
Welches ist das beste kostenlose Modell für langer kontext?
GLM 5.3 führt derzeit — kostenlos bei Together AI, NVIDIA Build. Danach folgen Mimo V2.6 Pro, Kimi K3, GLM 5.3 Flash.