Claude Haiku 5.5 vs Qwen 4 Max

Pick a route and service tier under each model. Prices, the calculator and the verdicts follow your choice.

Claude Haiku 5.5
Qwen 4 Max

At a glance

Generated from the rows below

Choose Claude Haiku 5.5 for

  • The lowest cost per typical request ($0.0005)

Choose Qwen 4 Max for

  • The only one that takes video
  • The longest output: 262,144 tokens

Overview

Context window
1Mtokens
1Mtokens
Max output
128Ktokens
262Ktokens (best)
Takes
Text, Image
Text, Image, Video
Returns
Text
Text
Routes on Phaseo
1 provider
1 provider
Released
7 Oct 2026
16 Nov 2026

Pricing

Prices on the route and tier chosen under each model

Anthropic · StandardAlibaba Cloud · Standard
Input
from $0.10/ 1M tokensInput ≤ 100K$0.10Input ≤ 100K · Region: US$0.11Input > 100K$0.50
$1.50/ 1M tokens
Output
from $0.50/ 1M tokensInput ≤ 100K$0.50Input ≤ 100K · Region: US$0.55Input > 100K$2.50
$4.50/ 1M tokens
Cache read
from $0.01/ 1M tokensInput ≤ 100K$0.01Input ≤ 100K · Region: US$0.011Input > 100K$0.05
$0.075/ 1M tokens
Cache write, 5 minutes
from $0.125/ 1M tokensInput ≤ 100K$0.125Input ≤ 100K · Region: US$0.1375Input > 100K$0.625
No write fee listed
Cache write, 1 hour
from $0.20/ 1M tokensInput ≤ 100K$0.20Input ≤ 100K · Region: US$0.22Input > 100K$1.00
No write fee listed
Cache write
No write fee listed
$1.875/ 1M tokens
Typical request2K tokens in, 500 out
$0.0005 (best)
$0.0053

What will it cost?

Estimate a month of requests on the route and tier chosen for each model

ModelPer requestPer month
Claude Haiku 5.5Anthropic · Standard$0.0005$13.50
Qwen 4 MaxAlibaba Cloud · Standard$0.0053$157.50

Uses the route and tier chosen under each model, over 30 days. Cached input is billed at the cache-read rate.

Performance

Live gateway traffic

Latency p50last 24 hours
No traffic in the last 5 days
1.93 s
Throughput p50last 24 hours
No traffic in the last 5 days
10.1 tps
Requests, last 24 hours
None in the last 24 hours
2

Activity

Tokens sent through Phaseo each day over the last 30 days

Tokens, last 30 days
8.98K
4.39K
Per day
5.04K
18 Oct16 Nov
4.39K
18 Oct16 Nov
Input
8.23K
4.36K
Output
751
15
Reasoning
0
15
Requests
32
2

Features

Features from the parameters Phaseo routes support

Tool calling
Yes
Yes
Structured outputs
Yes
Yes
Reasoning effort
Yes
Yes
Web search
Not listed
Yes
Prompt cachingon the chosen route
Yes
Yes
Quantisationon the chosen route
Not stated
Not stated

Benchmarks

Only results that two or more of the selected models report get a row

Compare at effort

Artificial Analysis Intelligence Index Independent

Higher is better · at Max effort · 1 of 2 models tested

Claude Haiku 5.536.5
Qwen 4 Max42.9

Where they rank

The five leaders on each capability index, then the compared models. Each model at its best-scoring setting.

Artificial Analysis Intelligence Index63 models ranked

  1. #1Claude Fable 5.264.6
  2. #2Gemini 4 Carbon60.0
  3. #3Claude Opus 5.559.1
  4. #4GPT-6 Astra56.9
  5. #5GPT-6.1 Sol54.6
  6. #16Qwen 4 Max42.9
  7. #18Claude Haiku 5.536.5

Epoch Capabilities Index165 models ranked

  1. #1Claude Fable 5.2173.4
  2. #2Gemini 4 Carbon169.8
  3. #3GPT-6 Astra169.0
  4. #4Claude Opus 5.5168.5
  5. #5GPT-6.1 Sol168.5
  6. #17Claude Haiku 5.5160.4
Cost
Artificial Analysis Evaluation Cost (USD)matched by effort level · lower is better
$339.47 (best)
$2,394

Lab-reported results come from each lab’s own launch materials and may use different settings, so none is marked best. Independent results come from third-party evaluators.

Providers

Where each model runs on Phaseo, with standard-tier prices per 1M tokens. Providers that serve more of these models come first.

—
$1.50 in · $4.50 out$10.00 per 1K requests Native web search requests1M context
$0.10 in · $0.50 out1M context · Standard, Batch
—
Also listed, not routed
15 providers
—

To price the whole page on one of these providers, pick it under the model’s name at the top.

Details

Knowledge cutoff
Not published
Not published
Status
Available
Available
Licence
Proprietary
Proprietary
Subscription plans
27 plansClaude Enterprise, Claude Free, Claude Max…
None mapped
API docs
Not published

Claude Haiku 5.5 vs Qwen 4 Max: side-by-side summary

Claude Haiku 5.5 and Qwen 4 Max are both available through the Phaseo API, so switching between them is a change of model id rather than a new integration.

Claude Haiku 5.5 is from Anthropic. It has a 1,000,000-token context window. On its cheapest route it costs $0.10 per 1M input tokens and $0.50 per 1M output tokens.

Qwen 4 Max is from Qwen. It has a 1,000,000-token context window. On its cheapest route it costs $1.50 per 1M input tokens and $4.50 per 1M output tokens.

Claude Haiku 5.5 model details, providers and pricingQwen 4 Max model details, providers and pricing

Claude Haiku 5.5 vs Qwen 4 Max at a glance

Claude Haiku 5.5 and Qwen 4 Max compared on price, context window and benchmark score
AttributeClaude Haiku 5.5Qwen 4 Max
LabAnthropicQwen
Input price per 1M tokens$0.10$1.50
Output price per 1M tokens$0.50$4.50
Context window1,000,000 tokens1,000,000 tokens
Max output128,000 tokens262,144 tokens
Artificial Analysis Intelligence Index36.5 (#18)42.9 (#16)

Prices are each model's cheapest standard route on Phaseo. Benchmark ranks use each model's best-scoring evaluated setting.

Frequently asked questions

Which is cheaper, Claude Haiku 5.5 or Qwen 4 Max?

Claude Haiku 5.5 is cheaper for both input and output tokens. Claude Haiku 5.5 costs $0.10 per 1M input tokens and $0.50 per 1M output tokens; Qwen 4 Max costs $1.50 and $4.50, each on its cheapest standard route on Phaseo.

Which has the longer context window, Claude Haiku 5.5 or Qwen 4 Max?

Both have a 1,000,000-token context window.

Which scores higher on the Artificial Analysis Intelligence Index, Claude Haiku 5.5 or Qwen 4 Max?

Qwen 4 Max scores higher on the Artificial Analysis Intelligence Index. Qwen 4 Max scores 42.9, ranked #16 of 63. Claude Haiku 5.5 scores 36.5, ranked #18 of 63.

Can I use Claude Haiku 5.5 and Qwen 4 Max through one API?

Yes. Both are served through the Phaseo API, so you can switch between them by changing the model id.

Help keep public data accurate

See something missing, incorrect, or out of date? Tell us about this page.