QwenOpen weightsReleased Feb 24, 2026

Qwen3.5-35B-A3B pricing, benchmarks and cost analysis

Compare Qwen guide

Intelligence#63 / 66

142.5

Epoch Capabilities Index (ECI)range 140.4 to 145.0

Speed

No speed data

Throughput not measured for this model

Cost#19 / 101

$212 / mo

In $0.140Out $1.00 per 1MCache discount 64%

Context#79 / 101

262K

Context window, tokens

Ranks are among the 101 models we track (intelligence = Epoch Capabilities Index by Epoch AI, CC BY 4.0; intelligence and speed: among those with data; cost: #1 is cheapest). Monthly cost uses the default workload of 2,000 input and 500 output tokens, 10,000 requests a day, 40% cached.

Comparison summary

Qwen3.5-35B-A3B costs $212 a month for the default workload of 2,000 input and 500 output tokens, 10,000 requests a day with 40% cached input. That is 76% below the median model ($873). It ranks #19 of 101 on cost, where #1 is the cheapest.

It scores 142.5 on the Epoch Capabilities Index (#63 of 66). Among the 8 other models within 2 points, 4 cost less, and it costs 2% less than their median ($216). The cheapest stable-priced model that scores at least as high is Qwen3.7 Flash (144.1) at $31.74 a month, 85% less.

Cached input is 64% cheaper than fresh input, a smaller discount than the 90% median. The weights are open, so other hosts can serve it and prices may differ from the provider's own. Its listed price changed on five or more days in the last month, usually because the cheapest host changed, so check the provider before relying on the figures above.

Technical specifications

Context window: 262K tokens
Input price: $0.140 / 1M
Cached input price: $0.050 / 1M
Output price: $1.00 / 1M
Cache discount: 64%
Batch API: Not listed
Open weights: Yes
Release date: Feb 24, 2026
Price volatility: Volatile (5+ changes in 30 days)

Benchmarks

Best score for this model on each leaderboard, ranked against the other models listed there.

Your monthly bill

Standard

$212 / month

Input: $50.40
Cached input: $12.00
Output: $150.00

Batch

No batch rates are listed for this model.

Price history

Oct 21, 2026Input $0.150 â†’ $0.140â–¼ 7%
Oct 14, 2026Input $0.080 â†’ $0.150â–² 87%
Oct 14, 2026Output $0.750 â†’ $1.00â–² 33%
Oct 14, 2026Cached input $0.040 â†’ $0.050â–² 25%
Oct 9, 2026Input $0.150 â†’ $0.080â–¼ 47%
Oct 9, 2026Output $1.00 â†’ $0.750â–¼ 25%
Oct 9, 2026Cached input $0.050 â†’ $0.040â–¼ 20%
Oct 7, 2026Input $0.080 â†’ $0.150â–² 87%
Oct 7, 2026Output $0.750 â†’ $1.00â–² 33%
Oct 7, 2026Cached input $0.040 â†’ $0.050â–² 25%
Oct 6, 2026Input $0.150 â†’ $0.080â–¼ 47%
Oct 6, 2026Output $1.00 â†’ $0.750â–¼ 25%
Oct 6, 2026Cached input $0.050 â†’ $0.040â–¼ 20%
Oct 2, 2026Input $0.163 â†’ $0.150â–¼ 8%
Oct 2, 2026Output $1.30 â†’ $1.00â–¼ 23%
Oct 2, 2026Cached input $0.016 â†’ $0.050â–² 213%
Sep 28, 2026Input $0.313 â†’ $0.163â–¼ 48%
Sep 28, 2026Output $1.25 â†’ $1.30â–² 4%
Sep 28, 2026Cached input $0.156 â†’ $0.016â–¼ 90%
Sep 20, 2026Input $0.163 â†’ $0.313â–² 92%
Sep 20, 2026Output $1.30 â†’ $1.25â–¼ 4%
Sep 20, 2026Cached input $0.016 â†’ $0.156â–² 875%
Sep 15, 2026Input $0.313 â†’ $0.163â–¼ 48%
Sep 15, 2026Output $1.25 â†’ $1.30â–² 4%
Sep 15, 2026Cached input $0.156 â†’ $0.016â–¼ 90%
Sep 6, 2026Input $0.080 â†’ $0.313â–² 291%
Sep 6, 2026Output $0.750 â†’ $1.25â–² 67%
Sep 6, 2026Cached input $0.008 â†’ $0.156â–² 1850%
Sep 5, 2026Input $0.250 â†’ $0.080â–¼ 68%
Sep 5, 2026Output $1.25 â†’ $0.750â–¼ 40%
Sep 5, 2026Cached input $0.250 â†’ $0.008â–¼ 97%

Cheaper alternatives

Models with stable pricing and a Epoch Capabilities Index within 2 points of Qwen3.5-35B-A3B (or higher) that cost less for the default workload.

Qwen3.7 Flash85% cheaper$31.74 a monthECI 144.1

Claude Haiku 5.547% cheaper$113 a monthECI 159.3

GPT-6 Luna47% cheaper$113 a monthECI 156.8

Gemma 4 31B36% cheaper$137 a monthECI 143.0