Is Qwen3 VL 235B A22B Thinking a good deal?
Whether anything in this catalogue beats Qwen3 VL 235B A22B Thinking on both quality and price, and what you give up if it does. A computation on the current catalogue, not an opinion.
13 undominated of 136 · Sep 23, 2026
As of Sep 23, 2026, Qwen3 VL 235B A22B Thinking is dominated for Balanced on LMArena. GLM 5.3 Flash scores 71.1 higher and costs 82% less, with a covering envelope. 13 of 136 rated, priced standard models are undominated.
Inspect model evidence Compare differences & requirements
GLM 5.3 Flash is both better and cheaper than Qwen3 VL 235B A22B Thinking: 71.1 points higher on LMArena and 82% less per million tokens, $1.06 cheaper at this mix.
LMArena Elo · higher is better
Scale starts at 1040 Elo
1400.8
Effective $/M · Balanced · lower is better
$1.30/M
LMArena Elo · higher is better
Scale starts at 1040 Elo
1471.9
Effective $/M · Balanced · lower is better
$0.237/M
Qwen3 VL 235B A22B Thinking takes text, image, returns up to 32,768 tokens from a 131,072-token context, and is listed by 2 sellers.
Compared against 136 rated, priced models on this lens: 16 models dominate it and give up nothing, 5 more dominate it but give something up. Qwen3 VL 235B A22B Thinking scores 1400.8 at $1.30 per million tokens for this mix.
Envelope-safe replacements
Each row scores at least as high, costs no more, and covers this model’s context, output, modalities, tools, and reasoning. A cheaper narrower model is not listed here.
| Model | LMArena | Effective $/M | You save |
|---|---|---|---|
| GLM 5.3 Flash | 1471.9 +71.1 | $0.237/M | 82% |
| Qwen3.7 Plus | 1454.2 +53.4 | $0.560/M | 57% |
| Kimi K2.5 | 1445.6 +44.8 | $0.900/M | 31% |
| Qwen3.8 27B | 1439.3 +38.5 | $1.06/M | 18% |
| Qwen3.5 397B A17B | 1438.3 +37.5 | $1.29/M | 1% |
| Qwen3.6 Plus | 1436.7 +35.9 | $0.731/M | 44% |
| Gemini 3.5 Flash Lite | 1435.5 +34.7 | $0.850/M | 35% |
| Gemma 4 26B A4B | 1434.5 +33.7 | $0.143/M | 89% |
| MiniMax M3 | 1433.5 +32.7 | $0.525/M | 60% |
| GPT-5.6 Luna | 1429.9 +29.1 | $0.450/M | 65% |
| MiMo-V2.5 | 1427.4 +26.6 | $0.175/M | 87% |
| Qwen3.5-122B-A10B | 1417.9 +17.1 | $0.715/M | 45% |
| Gemini 2.5 Flash | 1417.3 +16.5 | $0.850/M | 35% |
| Gemini 3.1 Flash Lite Preview | 1415.3 +14.5 | $0.563/M | 57% |
| Inkling Small | 1412.3 +11.5 | $0.637/M | 51% |
| Qwen3.5-27B | 1408.1 +7.3 | $0.536/M | 59% |
Higher score, lower price, named losses
Not a drop-in. The loss is why these are not a recommendation.
| Model | LMArena | Effective $/M | You give up |
|---|---|---|---|
| GLM 5.3 | 1475.1 | $1.29/M | image |
| GLM 5.2 | 1466.9 | $0.998/M | image |
| MiMo-V2.5-Pro | 1464.8 | $0.544/M | image |
| GLM 5 | 1446.3 | $0.930/M | image |
| DeepSeek V4 Pro 0423 | 1444.0 | $1.18/M | image |
The link keeps the model and mix, using the current catalogue. Monthly spend and switching cost stay in this browser tab and are left out of the link.
Saved decision references
Save the model, workload, catalogue date and capability-preservation rule in this browser. Spend, switching cost and bill contents are not saved. No account or notifications.
Constraint: replacements must preserve the model’s capabilities; any losses remain named trade-offs.
Historical decisions cannot be fully replayed from saved references: past prices, scores and capability evidence are not stored.