Compare / Qwen3.7-Max vs Qwen3.8-Max (0902)
Qwen3.7-Max vs Qwen3.8-Max (0902)
Which is cheaper, which is more capable, and which does better at each kind of work, from 14 results on 3 sources that measured both models.
Of the 14 results both models have, the better value on each. Results whose reported ranges overlap are too close to call.
Results as of 14 October 2026 (catalogue release 2026-10-14-7d7cf32c3f6d).
Qwen3.7-Max or Qwen3.8-Max (0902)? The short answer
Pick Qwen3.8-Max (0902) if quality matters most: it scores higher (Intelligence Index 45.4 against 29.5). Pick Qwen3.7-Max if cost matters more: it costs 26% less per token.
- Qwen3.7-Max is cheaper: $2.21 against $3.00 per million tokens, 26% less.
- Qwen3.8-Max (0902) scores higher on the Artificial Analysis Intelligence Index: 45.4 against 29.5.
- Qwen3.8-Max (0902) is newer: released 3 Aug 2026, against 19 May 2026 for Qwen3.7-Max.
- On third-party benchmarks, on the results both models have: Qwen3.7-Max does better for sticking to the facts; Qwen3.8-Max (0902) for agents and tool use and reasoning and knowledge.
Head to head
Every fact side by side
| Qwen3.7-Max | Qwen3.8-Max (0902) | |
|---|---|---|
| Price per million tokens | $1.48 in · $4.42 out | $2.00 in · $6.00 out |
| Blended (3 in : 1 out) | $2.21 | $3.00 |
| Cheapest host, blended | – | – |
| Intelligence Index | 29.5 | 45.4 |
| Spring Prompt overall | – | 46 |
| Output speed, tokens a second | – | 40 |
| Context window | 1,000,000 tokens | 1,000,000 tokens |
| Released | 19 May 2026 | 3 Aug 2026 (newer) |
| Weights | Closed | Closed |
| Developer based in | China | China |
Prices are the developer's list price where we have read it, otherwise the typical host on OpenRouter; see every model's price. Intelligence and speed from Artificial Analysis.
Which is better for what
Each model's rank among the models on sale today, for each kind of work, from our own benchmarks and licensed sources. Further right is better; the better rank is in green.
Every shared result
Each model's best result on every measure both have, grouped by benchmark, with the setting that produced it. Under each measure, every model that benchmark has measured is a grey tick, best to the right, so you can see where the two sit in the field. The headline measures of each benchmark come first; the rest are under "Show all". Green marks the better value; where a source reports ranges that overlap, neither is marked.
| Measure | Qwen3.7-Max | Qwen3.8-Max (0902) |
|---|---|---|
| Arena (formerly LMArena) reported by Arena (formerly LMArena) · 3 results | ||
| Confirmed task success · AgentIPS effect estimate, higher is better 39 · 20 of 48 measured | -0.09arena agent | 0.05arena agent |
| Praise over complaint · AgentIPS effect estimate, higher is better 39 · 20 of 48 measured | -0.09arena agent | 0.03arena agent |
| Steerability · AgentIPS effect estimate, higher is better 41 · 18 of 48 measured | -0.06arena agent | 0.03arena agent |
| Artificial Analysis reported by Artificial Analysis · 3 results | ||
| Artificial Analysis Intelligence Indexindex score, higher is better 54 · 15 of 244 measured | 29.5 | 45.4 |
| GPQA Diamond% of questions, higher is better 21 · 15 of 231 measured | 92.3% | 92.8% |
| Humanity's Last Exam% of questions, higher is better 40 · 25 of 243 measured | 40.5% | 43.1% |
| SimpleQA Verified (Epoch AI) reported by SimpleQA Verified (Epoch AI) · 1 result | ||
| Correct answers% of questions, higher is better 18 · 30 of 76 measured | 55.8% | 47.3%xhigh reasoning |
Show all 14 results (7 more)
| Measure | Qwen3.7-Max | Qwen3.8-Max (0902) |
|---|---|---|
| Arena (formerly LMArena) reported by Arena (formerly LMArena) · 1 result | ||
| Tool grounding · AgentIPS effect estimate, higher is better 34 · 43 of 48 measured | 0.00arena agent | -0.01arena agent |
| Artificial Analysis reported by Artificial Analysis · 6 results | ||
| Terminal-Bench 2.1% of tasks, higher is better 33 · 5 of 130 measured | 74.5% | 88.8% |
| Terminal-Bench 4.0% of tasks, higher is better 52 · 13 of 113 measured | 1.5% | 38.9% |
| τ-bench banking% of tasks, higher is better 71 · 7 of 121 measured | 11.8% | 47.8% |
| Artificial Analysis Coding Indexindex score, higher is better 34 · 10 of 131 measured | 66.0 | 76.2 |
| Long-context reasoning (AA-LCR)% of questions, higher is better 51 · 36 of 234 measured | 79.0% | 80.3% |
| SciCode% of problems, higher is better 53 · 37 of 114 measured | 49.5% | 52.1% |
Compare them with other models
Quick answers
Is Qwen3.7-Max better than Qwen3.8-Max (0902)?
Pick Qwen3.8-Max (0902) if quality matters most: it scores higher (Intelligence Index 45.4 against 29.5). Pick Qwen3.7-Max if cost matters more: it costs 26% less per token.
Which is cheaper, Qwen3.7-Max or Qwen3.8-Max (0902)?
Qwen3.7-Max is cheaper: $2.21 against $3.00 per million tokens, 26% less.
Which is better for agents and tool use, Qwen3.7-Max or Qwen3.8-Max (0902)?
Qwen3.8-Max (0902) ranks 9 of 166 models on sale for multi-step tasks with tools: support desks, coding agents, function calls, against 72 for Qwen3.7-Max.
Which is better for reasoning and knowledge, Qwen3.7-Max or Qwen3.8-Max (0902)?
Qwen3.8-Max (0902) ranks 18 of 200 models on sale for hard questions across science, maths and general knowledge, against 41 for Qwen3.7-Max.
Which is better for sticking to the facts, Qwen3.7-Max or Qwen3.8-Max (0902)?
Qwen3.7-Max ranks 31 of 153 models on sale for summarising without inventing things, and factual answers, against 39 for Qwen3.8-Max (0902).
Run Qwen3.7-Max and Qwen3.8-Max (0902) on your own prompt
Benchmarks aren't your data. Try both side by side in the playground, with the cost of every answer.