Gemma 4 26B A4B vs GPT-6.1 Terra
GPT-6.1 Terra averages 85.8% across five Vision Evals tasks and Gemma 4 26B A4B 54.1% across four (#2 and #52 of 65 models), so GPT-6.1 Terra leads overall.
- GPT-6.1 Terra wins all 4 tasks both ran, with its widest lead on object detection (84.7% vs 44.2%).
- Gemma 4 26B A4B is about half the cost: $0.0019 vs $0.0033 overall average cost per sample.
- GPT-6.1 Terra is faster: 10.1s vs 27.8s per sample.
Roboflow Vision Evals. Full breakdown below.
Compare Gemma 4 26B A4B vs GPT-6.1 Terra live
Run the same image across every model that supports a task and compare their outputs side-by-side.
These models don't share enough common tasks for a side-by-side demo. See the comparison table below for their capabilities.
Gemma 4 26B A4B vs GPT-6.1 Terra on Vision Evals
Roboflow Vision Evals. Full breakdown below.
Gemma 4 26B A4B vs GPT-6.1 Terra Comparison Table
Evals updated October 26, 2026Pricing updated October 27, 2026
| Property | Gemma 4 26B A4B | GPT-6.1 Terra |
|---|---|---|
| Organization | OpenAI | |
| Category | open | closed |
| Modality | multimodal | multimodal |
| Release Date | Apr 2026 | Oct 2026 |
| Context Window | 256K | 1.1M |
| Parameters | 25.2B | undisclosed |
| License | Apache 2.0 | Proprietary |
| Commercial use | free for commercial use | commercial use per vendor terms |
| Pricing per 1M tokens | ||
| Input $/1M | $0.090 | $1.00 |
| Output $/1M | $0.300 | $5.00 |
| Vision Tasks | ||
| Captioning | Demo | Supported |
| Classification | Demo | Supported |
| Object Detection | Demo | Supported |
| OCR | Demo | Supported |
| Visual Question Answering | Demo | Supported |
| Model Features | ||
Both support 3 more features | ||
Vision Evalsground-truth scores across 5 vision tasks, pooled at low effort | ||
| Overall | 54.1% 4/5 tasks | 85.8% |
| Quantizationsself-hosted | ||
| Avg cost / sample | $0.0019 | $0.0033 |
| Avg speed / sample | 27.84s | 10.05s |
| By task | ||
| Object Detection (low) | 44.2% ±0.7, Mean of 3 runs, range 43.5 to 44.8 | 84.7% ±0.1, Mean of 3 runs, range 84.7 to 84.8 |
| Object Detection (high) | – | 84.7% ±0.4, Mean of 3 runs, range 84.2 to 84.9 |
| Counting (low) | 43.2% ±2.0, Mean of 3 runs, range 41.9 to 46.0 | 80.2% ±2.0, Mean of 3 runs, range 78.4 to 82.4 |
| Counting (high) | – | 84.2% ±3.4, Mean of 3 runs, range 81.1 to 87.8 |
| Identification (low) | 81.3% ±3.1, Mean of 3 runs, range 78.1 to 84.4 | 96.9% ±0.0, Mean of 3 runs, range 96.9 to 96.9 |
| Identification (high) | – | 96.9% ±0.0, Mean of 3 runs, range 96.9 to 96.9 |
| OCR (low) | – | 81.9% |
| by category |
| |
| OCR (high) | – | 84.4% |
| by category |
| |
| Reasoning (low) | 47.7% ±2.0, Mean of 3 runs, range 45.0 to 49.0 | 85.4% ±1.7, Mean of 3 runs, range 83.4 to 86.8 |
| Reasoning (high) | – | 87.0% ±1.3, Mean of 3 runs, range 85.4 to 88.1 |
Gemma 4 26B A4B vs GPT-6.1 Terra: Overview
Gemma 4 26B A4B is the Mixture-of-Experts variant in Google's Gemma 4 family, with 25.2B total parameters but only 3.8B active per token. Built from the same Gemini 3 research as the 31B dense sibling and released as open weights under the Apache 2.0 license, it supports a 256K token context window with text and image input and configurable thinking mode. The "A4B" in the name refers to its approximately 4B active parameters. The MoE design makes it significantly faster at inference than the dense 31B, running nearly as fast as a 4B-parameter model while delivering roughly 97% of the dense model's quality.
For vision tasks, the 26B A4B shares the same multimodal capabilities as the 31B image understanding with variable aspect ratios and resolutions, and structured bounding box output for UI element detection. The tradeoff versus the 31B dense model is a small quality reduction in exchange for much faster inference and lower hardware requirements, fitting in 18GB of VRAM at 4-bit quantization. It ranked #6 among open models on the Arena AI text leaderboard at launch.
GPT-6.1 Terra is an OpenAI model released on October 26, 2026 for visual reasoning, computer use and writing. It accepts text and image inputs and produces text, with a 1,050,000-token context window and up to 128,000 output tokens. OpenAI lists applications including matching product photographs to catalog entries, interpreting technical drawings and operating application screens.
The API exposes six reasoning effort settings: none, low, medium, high, xhigh and max. Medium is the default. The model supports structured outputs and tool calling through the Responses API. Vision Evals reports its Low and High results for text understanding, counting, identification, reasoning and object detection, using native OpenAI API inference. Terra is a proprietary model; its weights are not released.
Frequently Asked Questions
On Roboflow's Vision Evals, GPT-6.1 Terra performed better. It scores higher on all 4 vision tasks and averages 85.8% (#2 of 65) against 54.1% (#52 of 65) for Gemma 4 26B A4B. These scores measure vision capabilities only; either model may perform differently on work outside this benchmark's tasks.
No. On the Vision Evals Object Detection benchmark at low effort, GPT-6.1 Terra leads with 84.7% against 44.2%. This is the widest gap between the two models across the benchmark's tasks.
Gemma 4 26B A4B is cheaper to run. Across Roboflow's Vision Evals task mix it averaged $0.0019 per sample against $0.0033. Actual costs depend on your image sizes, prompts, and output length.
GPT-6.1 Terra is faster. Across Roboflow's Vision Evals it averaged 10.1s per inference against 27.8s. Latency varies with image size, prompt length, and provider load, so treat these as relative rather than guaranteed figures.