Qwen3.7 Max

AlibabaProprietaryPreliminaryReleased May 19, 2026

On the arena.ai Agents leaderboard, Qwen3.7 Max has a Net Improvement of -7.12% (—, rank #45, 95% interval -8.36%–-5.89%). Qwen3.7 Max ranks #41 on the LMArena Text leaderboard (style control) with an Arena score of 1473.8 (±9.8, 3,901 votes). It has few votes so far, so this is a preliminary result and the rank may move. API price: $2.50 input / $7.50 output per 1M tokens. Context: 1M. Nearest rivals: GPT-6 Astra (1474.5) above and Gemini 3 Flash (1472.9) below.

Quick answer

Agents leaderboard
#45 of 37 · -7.12% success
Text (LMArena)
#41 · 1473.8
17NAS overall
#30 of 36
API price (in / out)
$2.50 / $7.50 per 1M tokens
Data updated
Nov 9, 2026 · synced daily

Sources: arena.ai, LMArena, Artificial Analysis, OpenRouter

Key metrics

Agent Net Improvement-7.12%43,146 sessions · #45
Arena score Preliminary1473.8±9.8LMArena #41 · 3,901 votes
Intelligence–Variant: max
Coding–Artificial Analysis
Writing (creative)1447.8±26.9LMArena creative writing #36 / 418
List price in / out$2.50 / $7.50USD per 1M tokens
Lowest route$1.48 / $4.43Cheapest OpenRouter route
Context1Mtokens
Output speed–AA measured median
TTFT / first answer–AA measured median

Rank history

LMArena rank, last 25 days
#38#41#43Oct 16Nov 9

#39 on Oct 16, #41 now; best #39 in this period.

Models ranked nearby

Sorted by Arena score
#ModelArenaIntel.Output $tok/s
30DeepSeek V4.1 Flash1474.9–$1.20–
31GLM-5.3 Flash1474.627.8$0.5062
32GPT-6 Astra1474.556.9$50.0051
33Qwen3.7 MaxThis page1473.8–$7.50–
34Gemini 3 Flash1472.9–$3.00–
35Claude Sonnet 4.61472.0–$15.00–
36Claude Sonnet 5.51471.954.5$10.00133

More from Alibaba

#ModelArenaIntel.Output $tok/s
20Qwen3.8 Max1484.233.7$6.0038
45Qwen3.6 Max1459.6–––
50Qwen3.7 Plus1455.4–$1.6058
63Qwen3 Max1423.4–––

Pricing

  • Official API list price: $2.50 input and $7.50 output per 1M tokens (USD).
  • At a typical 3:1 input-to-output ratio, the blended cost is about $3.75 per 1M tokens.
  • Among 68 models with an output price, it is the #35 cheapest.
  • The cheapest OpenRouter route charges $1.48 input / $4.43 output.

Sources & methodology

On the arena.ai Agents leaderboard, Qwen3.7 Max has a Net Improvement of -7.12% (—, rank #45, 95% interval -8.36%–-5.89%).

  • arena.ai Agents — agent Net Improvement, 95% interval, sessions and rank (via the public Hugging Face dataset, CC BY 4.0).
  • LMArena — Arena score (Text leaderboard with style control, the same view arena.ai shows by default), 95% confidence interval, votes and rank.
  • Artificial Analysis — Intelligence / Coding Index, output speed and latency (medians).
  • OpenRouter — Route prices, context length and listing date.

Arena scores use the Bradley-Terry model (not Elo); ± is the half-width of the 95% confidence interval. Latency figures are Artificial Analysis medians; for reasoning models, time to first token and to first answer include the thinking phase. Data syncs daily at 14:20 (UTC+8).

Limits: Arena scores come from anonymous users voting on chats, so they reflect conversational preference rather than agent or coding benchmarks; with few votes the interval is wide and ranks move. Prices and speeds change as providers update them.

FAQ

How does Qwen3.7 Max do on the Agents leaderboard?

On the arena.ai Agents leaderboard, Qwen3.7 Max has a Net Improvement of -7.12% (—, rank #45, 95% interval -8.36%–-5.89%).

What rank is Qwen3.7 Max right now?

Qwen3.7 Max is #33 on our leaderboard of 64 scored models (Arena score 1473.8); its raw LMArena rank is #41.

What does Qwen3.7 Max cost via API?

Input $2.50 / output $7.50 per 1M tokens (official API list price; the OpenRouter quote when no official price is available). At a typical 3:1 input-to-output ratio, the blended cost is about $3.75 per 1M tokens.

How close is Qwen3.7 Max to the models ranked next to it?

It is 0.7 points behind GPT-6 Astra (1474.5) and 0.9 points ahead of Gemini 3 Flash (1472.9); gaps smaller than the confidence interval (±9.8) are not significant.

Why is Qwen3.7 Max marked Preliminary?

It has only 3,901 votes on LMArena and a 95% confidence interval of ±9.8; with few votes the score can move noticeably as new votes arrive, so treat the rank as provisional. We mark models with fewer than 5,000 votes or ± above 10 as Preliminary.

Is Qwen3.7 Max open source? Can it be self-hosted?

Qwen3.7 Max is a closed model, available only via official APIs or hosted services.

What is the context length of Qwen3.7 Max?

1M (1,000,000 tokens, as listed on OpenRouter).

Qwen3.7 Max vs GPT-6 Astra: which is better?

On the LMArena Text leaderboard Qwen3.7 Max scores 1473.8 and GPT-6 Astra scores 1474.5 (a 0.7-point gap, within the confidence interval, so not significant). Agent task success: Qwen3.7 Max -7.12%, GPT-6 Astra 14.24%. On price, Qwen3.7 Max costs $7.50 and GPT-6 Astra $50.00 per 1M output tokens.

Is there a cheaper model close to Qwen3.7 Max?

Yes: Gemini 3.6 Flash scores 1482.7 on the Text leaderboard (Qwen3.7 Max: 1473.8) and costs $3.75 per 1M output tokens, about 50% less than Qwen3.7 Max ($7.50).

Where does the Qwen3.7 Max data come from and how often is it updated?

Agent success rates come from the arena.ai Agents leaderboard (CC BY 4.0), Text scores from LMArena, intelligence and speed from Artificial Analysis, and prices and context from official list prices and OpenRouter. 17NAS syncs daily at 06:20 UTC; this page was last updated Nov 9, 2026.

Related leaderboards