Gemma 4 26B A4B is 468.8× cheaper per million tokens (blended). GPT-6 Astra has the larger context window (1,050,000 vs 262,144 tokens, 4.0×). Gemma 4 26B A4B is open-weights; the other is not.
Gemma 4 26B A4B vs GPT-6 Astra
Live catalog fields. Best-in-row is highlighted.
Gemma 4 26B A4Bopen
Google · google__gemma-4-26b-a4b-it
GPT-6 Astra
OpenAI · openai__gpt-6-astra
Identity
| Field | google__gemma-4-26b-a4b-it | openai__gpt-6-astra |
|---|---|---|
| Provider | OpenAI | |
| Slug | google__gemma-4-26b-a4b-it | openai__gpt-6-astra |
| Status | live | live |
| Open weights | Yes | No |
| License | — | — |
| HuggingFace | google/gemma-4-26B-A4B-it | — |
Quality
| Field | google__gemma-4-26b-a4b-it | openai__gpt-6-astra |
|---|---|---|
| Quality index (0–100) | — | — |
Cost
| Field | google__gemma-4-26b-a4b-it | openai__gpt-6-astra |
|---|---|---|
| Input $ / 1M tokens | $0.03 | $10.00 |
| Cached input $ / 1M | $0.02 | $1.00 |
| Output $ / 1M tokens | $0.10 | $50.00 |
Context
| Field | google__gemma-4-26b-a4b-it | openai__gpt-6-astra |
|---|---|---|
| Max context tokens | 262,144 | 1,050,000 |
| Max output tokens | 32,768 | 128,000 |
Modalities
| Field | google__gemma-4-26b-a4b-it | openai__gpt-6-astra |
|---|---|---|
| Input | image, text, video | file, image, text |
| Output | text | text |