Compare / head-to-head
GLM-4.6Vvs
Qwen3-VL 235B-A22B Thinking
Qwen3-VL 235B-A22B Thinking leads 4 of 4 shared benchmarks. GLM-4.6V is 3.7x cheaper per token. Qwen3-VL 235B-A22B Thinking has the larger context window (256K tokens).
Benchmarks from cited public sources; pricing from official pages; status from official provider feeds.
Shared benchmarks
0 – 4
Qwen3-VL 235B-A22B Thinking leads
Cheaper per token
GLM-4.6V
3.7x cheaper, input + output
Larger context
Qwen3-VL 235B-A22B Thinking
256K tokens
Providers
2 providers
Z.ai · Qwen
GLM-4.6V
Z.ai · released 2025-12-08
textvisionvideo
Qwen3-VL 235B-A22B Thinking
Qwen · released 2025-09-23
textvisionvideo
Quality
Benchmark matrix
4 shared · 13 only GLM-4.6V · 13 only Qwen3-VL 235B-A22B Thinking| Benchmark | GLM-4.6V | Qwen3-VL 235B-A22B Thinking | Δ | Edge |
|---|---|---|---|---|
| Reported by both · 4 | ||||
| LMArena Elo | 1378.7 ↗ | 1395 ↗ | -16.3 | Qwen3-VL 235B-A22B Thinking |
| LMArena Vision (LMArena) | 1161.5 ↗ | 1188 ↗ | -26.5 | Qwen3-VL 235B-A22B Thinking |
| MMMU_Pro | 66% ↗ | 69.3% ↗ | -3.3 pt | Qwen3-VL 235B-A22B Thinking |
| OSWorld | 37.2% ↗ | 38.1% ↗ | -0.9 pt | Qwen3-VL 235B-A22B Thinking |
| Only GLM-4.6V reports · 13 | ||||
| AndroidWorld | 57% ↗ | not reported | — | — |
| ChartQAPro | 65.5% ↗ | not reported | — | — |
| CharXiv_Val-Reasoning | 63.2% ↗ | not reported | — | — |
| Design2Code | 88.6% ↗ | not reported | — | — |
| MathVista | 85.2% ↗ | not reported | — | — |
| MMBench V1.1 | 88.8% ↗ | not reported | — | — |
| MMBrowseComp | 7.6% ↗ | not reported | — | — |
| MMLongBench-Doc | 54.9% ↗ | not reported | — | — |
| MMMU (Val) | 76% ↗ | not reported | — | — |
| MMStar | 75.9% ↗ | not reported | — | — |
| OCRBench | 86.5% ↗ | not reported | — | — |
| VideoMMMU | 74.7% ↗ | not reported | — | — |
| WebVoyager | 81% ↗ | not reported | — | — |
| Only Qwen3-VL 235B-A22B Thinking reports · 13 | ||||
| MMLU-Pro | not reported | 83.8% ↗ | — | — |
| AIME25 | not reported | 89.7% ↗ | — | — |
| BFCL | not reported | 71.9% ↗ | — | — |
| HLE | not reported | 13.6% ↗ | — | — |
| HMMT25 | not reported | 77.4% ↗ | — | — |
| IFEval | not reported | 88.2% ↗ | — | — |
| LCB v6 | not reported | 70.1% ↗ | — | — |
| MathVista (mini) | not reported | 85.8% ↗ | — | — |
| MMLU-Redux | not reported | 93.7% ↗ | — | — |
| MMMU (val) | not reported | 80.6% ↗ | — | — |
| SimpleQA | not reported | 44.4% ↗ | — | — |
| SuperGPQA | not reported | 64.3% ↗ | — | — |
| VideoMME (w/o sub) | not reported | 79% ↗ | — | — |
Scores tagged "epoch" or "matharena" are independent runs, used only where the lab has not published its own; ⚠ marks rows MathArena flags as released after the competition. Higher is better on every row. Δ is GLM-4.6V minus Qwen3-VL 235B-A22B Thinking in the benchmark's own unit. "Not reported" means the lab has not published that figure; it is not a zero. ↗ opens the source.
Specs & pricing
Side by side
Official pricing pages and model cards| Spec | GLM-4.6V | Qwen3-VL 235B-A22B Thinking | Edge |
|---|---|---|---|
| Context window | 128K tokens | 256K tokens | Qwen3-VL 235B-A22B Thinking |
| Max output | 32K tokens | — | — |
| Input price / 1M | $0.3 ↗ | $0.4 ↗ | GLM-4.6V |
| Output price / 1M | $0.9 ↗ | $4 ↗ | GLM-4.6V |
| Cached input / 1M | $0.05 ↗ | — | — |
| Input + output / 1M Lower is cheaper. List prices; batch, tool and regional fees excluded. | $1.2 | $4.4 | GLM-4.6V |
| Modalities | text · vision · video | text · vision · video | Tie |
| Released | 2025-12-08 | 2025-09-23 | — |
| Cited benchmark scores | 17 | 17 | — |
Reliability
Provider status
Live from /statusMore matchups
GLM-4.6V vs …
Models sharing the most benchmarksMore matchups
Qwen3-VL 235B-A22B Thinking vs …
Models sharing the most benchmarksBuilt by Respan
Which one wins on your data?
Public benchmarks are a starting point. Run GLM-4.6V and Qwen3-VL 235B-A22B Thinking on your own prompts with Respan evals, or route to either through one gateway key with automatic failover.