The Known Good Updated 4 Sep 2026 Subscribe
Qwen: Qwen3.6 Max Preview vs Qwen: Qwen3.7 Max ·Alibaba vs Alibaba

Qwen: Qwen3.6 Max Preview vs Qwen: Qwen3.7 Max — head-to-head comparison

Qwen: Qwen3.6 Max Preview profile Qwen: Qwen3.7 Max profile All comparisons
Qwen: Qwen3.6 Max Preview — Alibaba
Quality
#12 of 65
89based on 3 of 7 evaluations
Known Good Index
Speed
#176 of 338
46.4
Output tokens per second
Price
#323 of 442
$2.31
Blended per 1M tokens
In $1.03 · Out $6.16
Cache hit price
#187 of 324
$0.13
per 1M tokens (−87%)
Context
#220 of 476
262k
tokens (~400 pages)
Qwen: Qwen3.7 Max — Alibaba
Quality
#6 of 65
93based on 3 of 7 evaluations
Known Good Index
Speed
#114 of 338
63.6
Output tokens per second
Price
#319 of 442
$2.21
Blended per 1M tokens
In $1.48 · Out $4.42
Cache hit price
#276 of 324
$0.29
per 1M tokens (−80%)
Context
#149 of 476
1.00M
tokens (~1,500 pages)
Summary

On the Known Good Index, Qwen: Qwen3.6 Max Preview scores 89 and Qwen: Qwen3.7 Max scores 93 — a 3-point lead for Qwen: Qwen3.7 Max. They share 3 published evaluations, with Qwen: Qwen3.6 Max Preview ahead on 0 and Qwen: Qwen3.7 Max ahead on 3.

Blended pricing is $2.31 per 1M tokens for Qwen: Qwen3.6 Max Preview and $2.21 for Qwen: Qwen3.7 Max, within 10% of each other.

Throughput is 46 tokens per second for Qwen: Qwen3.6 Max Preview and 64 for Qwen: Qwen3.7 Max, with time to first token at 1.9s and 0.9s. Qwen: Qwen3.6 Max Preview accepts 262k tokens of context against 1.00M.

Qwen: Qwen3.6 Max Preview is proprietary; Qwen: Qwen3.7 Max is proprietary. 1 tracked host serve one or both, and 1 serve both.

Generated from tracked data on 4 Sep 2026 · how these comparisons are computed →

At a glance
MetricQwen: Qwen3.6 Max PreviewQwen: Qwen3.7 Max
Known Good Index 89 based on 3 of 7 evaluations 93 based on 3 of 7 evaluations
Blended $/1M $2.31 $2.21
Output tok/s 46 64
Context window 262k 1.00M
Providers 1 1
Jump to section

Quality

Known Good Index and the capability indices, side by side

Index scores

Each index is min-max normalised across tracked models and scaled 0–100 from published evaluations. We do not run these evaluations.

MetricQwen: Qwen3.6 Max PreviewQwen: Qwen3.7 Max
Known Good Index 89 93
Openness Index 5 5
The Known Good

Quality Breakdown

Every evaluation both models report a published score for

Evaluations both models report

Bar = Qwen: Qwen3.6 Max Preview · marker = Qwen: Qwen3.7 Max · scores ingested from Epoch AI (CC-BY 4.0)

GPQA Diamond87 / 91
Mock AIME 2024–2591 / 96
SWE-bench Verified77 / 77

Right-hand figures read Qwen: Qwen3.6 Max Preview / Qwen: Qwen3.7 Max. Qwen: Qwen3.6 Max Preview leads on 0 of 3, Qwen: Qwen3.7 Max on 3.

The Known Good

Arena Elo

Human preference rating from blind pairwise votes

Arena ratings

Elo from blind pairwise votes, ingested from Arena. Higher is better.

MetricQwen: Qwen3.6 Max PreviewQwen: Qwen3.7 Max
Text Arena 1,446 —
Text Arena (style controlled) 1,460 —
The Known Good

Price & Cost

Published API pricing by token type

Published pricing

USD per 1M tokens. Blended is a 3:1 input:output mix. Live from OpenRouter, cross-checked against models.dev. Lower is better.

MetricQwen: Qwen3.6 Max PreviewQwen: Qwen3.7 Max
Input $/1M $1.03 $1.48
Output $/1M $6.16 $4.42
Blended $/1M $2.31 $2.21
Cache read $/1M $0.130 $0.295
The Known Good
We publish blended price rather than cost-per-task. Cost-per-task requires token counts from an evaluation suite we do not run. Blended price is computed from published rates and is fully reproducible — see the pricing methodology.

Context Window

Maximum input accepted and maximum output emitted

Context and output limits

Provider-declared limits, from metadata via models.dev and OpenRouter. Higher is better.

MetricQwen: Qwen3.6 Max PreviewQwen: Qwen3.7 Max
Context window 262k 1.00M
Max output tokens 66k 131k
The Known Good

Speed & Latency

Mean throughput and time to first token across hosting providers, over successful measurements from the last 48 hours

Throughput and latency

Means across all tracked providers over successful measurements from the last 48 hours, derived from OpenRouter endpoint telemetry.

MetricQwen: Qwen3.6 Max PreviewQwen: Qwen3.7 Max
Output tokens/s 46 64
Time to first token 1.9s 0.9s
The Known Good

Providers

Which hosts serve each model, and at what price and speed

1 host · 1 serve both
Full provider benchmarks →
Provider Qwen: Qwen3.6 Max Preview blendedQwen: Qwen3.6 Max Preview tok/s Qwen: Qwen3.7 Max blendedQwen: Qwen3.7 Max tok/s
Alibaba · first party $2.31 49 $2.21 52

Per-host pricing and telemetry from OpenRouter. A dash means that host does not serve that model.

The Known Good
Sources for this page
Pricing & providers — OpenRouter Metadata — models.dev Benchmarks — Epoch AI (CC-BY 4.0)Elo — Arena Full attribution →