Across 1 comparable benchmark, Phi-4-Reasoning-Vision-15B has better reported results on 1 and worse on 0.
Limited evidence1 of 1 shared result countedHow we compare
Price basis: Listed rates vs listed rates. The price section below shows the applicable periods.
| Fact | Phi-4-Reasoning-Vision-15B | Qwen-VL-MAX |
|---|---|---|
| Context window | 16K tokens | 2K tokens |
| Input / 1M | — | — |
| Output / 1M | — | — |
| Blended (4:1) | — | — |
| Status | Available | Deprecated |
| Released | March 4, 2026 | January 25, 2024 |
Blended is 4:1 input to output with 90% cache hits at listed rates — set your own mix.
Mean of 1 shared benchmark grouped by capability on a 0–100 scale. How averages are built
Phi-4-Reasoning-Vision-15B ahead on 1 of 1 counted
No shared comparable results for Coding, Abstract reasoning, Mathematics, Science & health, Computer & tool use, Research & search, Long context, Professional work & productivity, Cybersecurity.
Published scores, with matching modes where available. Gaps show Phi-4-Reasoning-Vision-15B relative to Qwen-VL-MAX. Reading the results
| Benchmark | Phi-4-Reasoning-Vision-15B | Qwen-VL-MAX | Gap |
|---|---|---|---|
| MMMUVisual Understanding · Accuracy | 55.0Thinking · no tools | 51.4≠Reported best | +3.6 pts#49 vs #51 of 62 |
Visual Understanding · Accuracy
A dash means no result is available.
| Benchmark | Phi-4-Reasoning-Vision-15B | Qwen-VL-MAX |
|---|---|---|
| ScreenSpot-v2 | 88.3 | — |
| AI2D | 84.7 | — |
| ChartQA | 82.9 | — |
| HallusionBench | 63.9 | — |
| MathVerse (testmini) | 53.1 | — |
| MathVision (testmini) | 36.2 | — |
| MathVista (testmini) | 74.1 | — |
| MMStar | 63.9 | — |
| OCRBench | 75.6 | — |
Time-of-day providers use peak-hour rates; fixed-price providers use standard rates. Off-peak discounts remain in the published details.
Drag the handle above the bar for cache hits, the one below it for the input : output mix · 5% steps
Blended price = effective input rate at 90% cache hits × 80% + output rate × 20%
A blended cost needs a published input and output price on both sides, and one of these two does not have them yet.
Phi-4-Reasoning-Vision-15B: Listed text rates; pricing tier not confirmed.
Qwen-VL-MAX: Listed text rates; pricing tier not confirmed.
| Price / 1M tokens | Phi-4-Reasoning-Vision-15B | Qwen-VL-MAX | Ratio |
|---|---|---|---|
| Input | — | — | — |
| Output | — | — | — |
| Cached input | — | — | — |
| Cache write | — | — | — |
| Blended | — | — | — |
Estimates use the selected token mix. Cache writes, tool fees and long-context surcharges are excluded.
Lower prices are highlighted. The cache hit rate applies to input tokens only; models without a published cached-input rate use their normal input price for those tokens.| Attribute | Phi-4-Reasoning-Vision-15B | Qwen-VL-MAX |
|---|---|---|
| Availability | Available | Deprecated |
| Provider | Microsoft | Alibaba |
| Model type | Reasoning model | Multimodal model |
| First release | March 4, 2026 | January 25, 2024 |
| Context window | 16K tokens | 2K tokens |
| Total parameters | 15B | — |
| Architecture | Mixture of Experts (MoE) | Undisclosed |
| Documented capabilities | Reasoning, OCR, Computer use | Multilingual |
| Input modalities | Text, Image | — |
| Output modalities | Text | — |
| Weights | Open weights | Closed / API only |
| Licence | MIT License | — |
| Commercial use | Commercial use allowed | Proprietary |
These are reported-score comparisons, not controlled head-to-head tests.
What people found comparing these models, or using either one. Comments also appear on the models' own pages.
Used Phi-4-Reasoning-Vision-15B or Qwen-VL-MAX? Sign in with Google to share what worked, what did not, and which API or setup you tested.
No community comments on either model yet.
Alternatives with the most benchmark overlap for this matchup.