Grok 4.7 vs Qwen3.8 27B
Compare Grok 4.7 and Qwen3.8 27B side-by-side.
Compare Grok 4.7 vs Qwen3.8 27B 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.
Models in this comparison
Grok 4.7 vs Qwen3.8 27B on Vision Evals
Grok 4.7 scores higher on 4 of the six Vision Evals tasks.
The widest gap is Object Detection, where Qwen3.8 27B leads 65.7% to 40.4%.
Overall, Grok 4.7 averages 71.9% (#22 of 54) against 74.7% (#17 of 54) for Qwen3.8 27B.
Qwen3.8 27B is both cheaper ($0.0009 vs $0.012 per sample) and faster (18.0s vs 23.6s per sample).
Grok 4.7 vs Qwen3.8 27B Comparison Table
Evals updated September 22, 2026Pricing updated September 22, 2026
| Property | Grok 4.7 | Qwen3.8 27B |
|---|---|---|
| Organization | SpaceXAI | Qwen |
| Category | closed | open |
| Modality | — | multimodal |
| Release Date | Sep 2026 | Aug 2026 |
| Context Window | 500K | 262K |
| Parameters | 27.78B | |
| License | Proprietary | Apache 2.0 |
| Pricing per 1M tokens | ||
| Input $/1M | $1.60 | $0.420 |
| Output $/1M | $4.80 | $3.00 |
| Vision Tasks | ||
| Captioning | Demo | |
| Chart Question Answering | ||
| Classification | Demo | |
| Document Question Answering | ||
| Image Tagging | ||
| Multi-Label Classification | ||
| Object Detection | Demo | |
| OCR | Demo | |
| Vision Language | ||
| Visual Question Answering | Demo | |
| Model Features | ||
| Foundation Vision | ||
| LLMs with Vision Capabilities | ||
| Multimodal Vision | ||
Vision Evalsground-truth scores across 6 vision tasks, pooled at low effort | ||
| Overall | 71.9% | 74.7% |
| Quantizationsself-hosted | ||
| Avg cost / sample | $0.012 | $0.0009 |
| Avg speed / sample | 23.55s | 17.99s |
| By task | ||
| Object Detection (low) | 40.4% ±0.6, Mean of 3 runs, range 39.8 to 41.0 | 65.7% ±1.0, Mean of 3 runs, range 64.6 to 66.5 |
| Object Detection (high) | 41.2% ±1.6, Mean of 3 runs, range 39.6 to 42.8 | 66.1% ±1.4, Mean of 3 runs, range 64.9 to 67.8 |
| Counting (low) | 61.7% ±1.3, Mean of 3 runs, range 60.8 to 63.5 | 64.9% ±4.1, Mean of 3 runs, range 60.8 to 68.9 |
| Counting (high) | 60.8% ±1.3, Mean of 3 runs, range 59.5 to 62.2 | 68.0% ±2.0, Mean of 3 runs, range 66.2 to 70.3 |
| Identification (low) | 87.5% ±3.1, Mean of 3 runs, range 84.4 to 90.6 | 85.4% ±4.7, Mean of 3 runs, range 81.3 to 90.6 |
| Identification (high) | 80.2% ±1.6, Mean of 3 runs, range 78.1 to 81.3 | 87.5% ±3.1, Mean of 3 runs, range 84.4 to 90.6 |
| OCR (low) | 92.6% ±0.7, Mean of 3 runs, range 92.1 to 93.4 | 92.2% ±1.2, Mean of 3 runs, range 91.1 to 93.4 |
| OCR (high) | 93.5% ±0.3, Mean of 3 runs, range 93.1 to 93.8 | 91.5% ±1.4, Mean of 3 runs, range 90.1 to 92.9 |
| Data Extraction (low) | 84.9% ±2.6, Mean of 3 runs, range 82.5 to 87.6 | 78.0% ±1.0, Mean of 3 runs, range 77.3 to 79.4 |
| Data Extraction (high) | 87.6% ±1.5, Mean of 3 runs, range 86.6 to 89.7 | 80.8% ±1.0, Mean of 3 runs, range 79.4 to 81.4 |
| Reasoning (low) | 64.2% ±2.3, Mean of 3 runs, range 62.3 to 66.9 | 62.0% ±2.0, Mean of 3 runs, range 60.3 to 64.2 |
| Reasoning (high) | 66.9% ±1.3, Mean of 3 runs, range 65.6 to 68.2 | 66.0% ±0.7, Mean of 3 runs, range 65.6 to 66.9 |
Grok 4.7 vs Qwen3.8 27B: Overview
Grok 4.7 is a proprietary model from SpaceXAI, released on September 21, 2026. It accepts text and images as input and returns text. It extends Grok 4.6 and is listed at the same API price.
Its Vision Evals scores are on the leaderboard. Running it in the Playground is not available yet, because the inference workflow is not ready.
Qwen3.8-27B is a dense vision-language model of roughly 27.78 billion parameters from Alibaba's Qwen team, built on the architectural foundation established by the Qwen3.5 generation. It takes interleaved text, image, and video input through an integrated vision encoder and produces text, covering document and chart understanding, visual question answering, front-end code generation from visual references, and agentic computer-use workflows. The decoder is a hybrid stack of 64 layers that repeats a pattern of three Gated DeltaNet linear-attention blocks, each followed by a feed-forward block, then one grouped-query full-attention block, so only a quarter of the layers maintain a conventional key-value cache. Native context length is 262,144 tokens.
Post-training targets long-horizon agentic work, and the model exposes per-request thinking control that lets callers switch reasoning traces on or off and choose a reasoning effort level. Multi-token prediction weights are trained alongside the main network for speculative decoding. Qwen reports improvements over Qwen3.6-27B on agentic and multimodal evaluations including Terminal-Bench 2.1, OSWorld-Verified, and SWE-bench Multimodal, with additional results on MathVision, CharXiv, and an internal Vision2Web suite.
Frequently Asked Questions
On Roboflow's Vision Evals, Grok 4.7 performed better. It scores higher on 4 of the six vision tasks and averages 71.9% (#22 of 54) against 74.7% (#17 of 54) for Qwen3.8 27B. 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, Qwen3.8 27B leads with 65.7% against 40.4%. This is the widest gap between the two models across the benchmark's tasks.
Qwen3.8 27B is cheaper to run. Across Roboflow's Vision Evals task mix it averaged $0.0009 per sample against $0.012. Actual costs depend on your image sizes, prompts, and output length.
Qwen3.8 27B is faster. Across Roboflow's Vision Evals it averaged 18.0s per inference against 23.6s. Latency varies with image size, prompt length, and provider load, so treat these as relative rather than guaranteed figures.