Alibaba Cloud: Models Intelligence, Performance & Price
This analysis is intended to support you in choosing the best model provided by Alibaba Cloud for your use-case.
Most Intelligent
Qwen3.8 Max (0902)Qwen3.8 MaxQwen3.8 2.4T A95BQwen3.8-Flash-NextQwen3.8 27B (xhigh)Intelligence Index
Total 27 models
Fastest
Qwen3.5 Omni FlashQwen3.7 MaxQwen3.5 35B A3B (non-reasoning)Qwen3.5 35B A3BQwen3.6 35B A3B (non-reasoning)Output speed
Total 27 models
Lowest Price
Qwen3.8-Flash-NextQwen3.5 Omni FlashQwen3.7 PlusQwen3 Omni 30B A3BQwen3 Omni 30B A3BBlended price (per 1M tokens)
Total 27 models
Alibaba Cloud offers 27 models, each with different intelligence, performance, and pricing characteristics. Below is a comparison of the key metrics across models.
- For intelligence, the top models on Alibaba Cloud are Qwen3.8 Max (0902) (45), Qwen3.8 Max (40), and Qwen3.8 2.4T A95B (40).
- For output speed, the fastest models are Qwen3.5 Omni Flash (235 t/s), Qwen3.7 Max (202 t/s), and Qwen3.5 35B A3B (non-reasoning) (165 t/s). Speed varies significantly across models, with a 61% difference between the fastest and slowest.
- For latency, Qwen3.5 Omni Flash (1.79s), Qwen3 Omni 30B A3B (1.95s), and Qwen3.5 35B A3B (non-reasoning) (2.08s) offer the lowest time to first answer token.
- For pricing, Qwen3.8-Flash-Next ($0.09), Qwen3.5 Omni Flash ($0.17), and Qwen3.7 Plus ($0.30) offer the lowest blended prices per 1M tokens. Prices vary up to 3.7x across models.
- For context window size, Qwen3.8 Max (1M), Qwen3.8-Flash-Next (1M), and Qwen3.8 27B (xhigh) (1M) support the largest context windows on Alibaba Cloud.
Intelligence Evaluations
Artificial Analysis Intelligence Index
Evaluations
Agentic knowledge work, Elo
Knowledge
1 - hallucination rate
Agentic SaaS workflows
Agentic real-world work tasks, Elo
Reasoning & knowledge
Professional document reasoning, All-pass
Agentic coding & terminal use
Long context reasoning
Coding
Physics reasoning
Agentic scientific research workflows in a terminal
Agentic software engineering in open-source repositories
Quantitative analysis on spreadsheets & documents
Legal agentic work, Hallucination-Gated All-Pass Rate
Agentic business operations
Agentic interactive 3D experiences, Elo
Product and brand videos designed in code, Elo
Kubernetes incident root-cause analysis
Medical long context reasoning
Visual reasoning
Rich, varied, well-crafted language
Well-structured, scannable, answer-first
Truthful, acknowledges uncertainty, non-sycophantic
Use of filler and generic language
Intelligence Index vs. Price
Context Window
Context Window
Pricing
Intelligence Index vs. Price
Performance Summary
Output Speed vs. Price
Speed
Measured by Output Speed (tokens per second)
Output Speed
Latency
Measured by Time (seconds) to First Token
Latency: Time To First Answer Token
Cache Behaviour
Cache Hit Rate
Cost per Task vs. Cache Hit Rate
End-to-End Response Time
Seconds to output 500 tokens, calculated based on time to first token, 'thinking' time for reasoning models, and output speed
End-to-End Response Time vs. Price
Further Analysis | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
Qwen3.8 Max (0902) | 984k | Proprietary | 45 | $5.41 | 41 | 2.91 | 63.15 | 48.20 | |||
Qwen3.8 Max | 1M | Proprietary | 40 | $2.67 | 41 | 2.70 | 63.89 | 48.95 | |||
Qwen3.8 2.4T A95B | 262k | Open | 40 | $2.16 | 40 | 3.04 | 65.01 | 49.57 | |||
Qwen3.8-Flash-Next | 1M | Open | 40 | $0.37 | 52 | 2.06 | 49.76 | 38.16 | |||
Qwen3.8 27B (xhigh) | 1M | Open | 34 | $1.15 | 44 | 3.87 | 61.10 | 45.78 | |||
Qwen3.7 Max | 991k | Proprietary | 29 | $0.93 | 202 | 2.19 | 16.58 | 11.91 | |||
Qwen3.8 27B (medium) | 1M | Open | 28 | $1.27 | 46 | 3.92 | 58.08 | 43.33 | |||
Qwen3.6 Plus | 1M | Proprietary | 27* | -- | 55 | 2.17 | 112.67 | 101.37 | |||
Qwen3.8 27B (low) | 1M | Open | 26 | $1.17 | 47 | 3.95 | 56.83 | 42.31 | |||
Qwen3.7 Plus | 1M | Proprietary | 25 | $0.22 | 57 | 2.30 | 46.13 | 35.06 | |||
Qwen3.5 27B | 262k | Open | 23* | -- | 77 | 5.66 | 38.08 | 25.93 | |||
Qwen3.5 397B A17B (non-reasoning) | 262k | Open | 21* | -- | 59 | 2.32 | 10.80 | -- | |||
Qwen3.6 27B | 262k | Open | 21 | $0.62 | 57 | 3.61 | 111.41 | 99.07 | |||
Qwen3.5 Omni Plus | 256k | Proprietary | 20* | -- | 119 | 2.10 | 6.30 | -- | |||
Qwen3.8 27B (non-reasoning) | 1M | Open | 20 | $2.49 | 48 | 3.95 | 14.27 | -- | |||
Qwen3.6 27B (non-reasoning) | 262k | Open | 20* | -- | 58 | 3.61 | 12.20 | -- | |||
Qwen3.5 27B (non-reasoning) | 262k | Open | 19* | -- | 81 | 5.65 | 11.83 | -- | |||
Qwen3.5 35B A3B | 262k | Open | 19* | -- | 150 | 2.07 | 18.69 | 13.30 | |||
Qwen3.5 397B A17B | 262k | Open | 18 | $0.47 | 58 | 2.26 | 65.86 | 54.97 | |||
Qwen3.6 35B A3B | 229k | Open | 18 | $0.48 | 133 | 2.07 | 46.47 | 40.64 | |||
Qwen3.5 122B A10B (non-reasoning) | 262k | Open | 18* | -- | 145 | 2.30 | 5.76 | -- | |||
Qwen3.5 122B A10B | 262k | Open | 16 | $0.32 | 131 | 2.31 | 21.36 | 15.24 | |||
Qwen3.6 35B A3B (non-reasoning) | 262k | Open | 15* | -- | 146 | 2.08 | 5.49 | -- | |||
Qwen3.5 35B A3B (non-reasoning) | 262k | Open | 15* | -- | 164 | 2.08 | 5.13 | -- | |||
Qwen3.5 Omni Flash | 256k | Proprietary | 12* | -- | 237 | 1.79 | 3.90 | -- | |||
Qwen3 Omni 30B A3B | 65.5k | Open | 8* | -- | 103 | 1.90 | 26.22 | 19.46 | |||
Qwen3 Omni 30B A3B | 65.5k | Open | 6* | -- | 96 | 1.94 | 7.13 | -- | |||
Key definitions
Frequently Asked Questions
Common questions about Alibaba Cloud
Alibaba Cloud offers 27 models that we track: Qwen3.8 Max (0902), Qwen3.8 Max, Qwen3.8 2.4T A95B, Qwen3.8-Flash-Next, Qwen3.8 27B (xhigh), Qwen3.7 Max, Qwen3.8 27B (medium), Qwen3.6 Plus, Qwen3.8 27B (low), Qwen3.7 Plus, Qwen3.5 27B, Qwen3.5 397B A17B (non-reasoning), Qwen3.6 27B, Qwen3.5 Omni Plus, Qwen3.8 27B (non-reasoning), Qwen3.6 27B (non-reasoning), Qwen3.5 27B (non-reasoning), Qwen3.5 35B A3B, Qwen3.5 397B A17B, Qwen3.6 35B A3B, Qwen3.5 122B A10B (non-reasoning), Qwen3.5 122B A10B, Qwen3.6 35B A3B (non-reasoning), Qwen3.5 35B A3B (non-reasoning), Qwen3.5 Omni Flash, Qwen3 Omni 30B A3B, and Qwen3 Omni 30B A3B.
The most intelligent model available on Alibaba Cloud is Qwen3.8 Max (0902) with an Intelligence Index score of 45.
The fastest model on Alibaba Cloud by output speed is Qwen3.5 Omni Flash at 235.3 tokens per second.
The model with the lowest time to first answer token on Alibaba Cloud is Qwen3.5 Omni Flash at 1.79s. Lower latency means faster initial response time.
The most affordable model on Alibaba Cloud by blended price is Qwen3.8-Flash-Next at $0.09 per 1M tokens (7:2:1 cache hit/input/output ratio).
Prices on Alibaba Cloud vary up to 18x across models, from $0.09 per 1M tokens for Qwen3.8-Flash-Next to $1.60 per 1M tokens for Qwen3.7 Max.
Yes, Alibaba Cloud offers an OpenAI-compatible API, making it easy to switch from OpenAI or use existing OpenAI SDK integrations.
Yes, all 27 models on Alibaba Cloud support JSON mode for structured output.
Yes, all 27 models on Alibaba Cloud support function calling (tool use).
Yes, Alibaba Cloud offers 17 reasoning models: Qwen3.8 Max (0902), Qwen3.8 Max, Qwen3.8 2.4T A95B, Qwen3.8-Flash-Next, Qwen3.8 27B (xhigh), Qwen3.7 Max, Qwen3.8 27B (medium), Qwen3.6 Plus, Qwen3.8 27B (low), Qwen3.7 Plus, Qwen3.5 27B, Qwen3.6 27B, Qwen3.5 35B A3B, Qwen3.5 397B A17B, Qwen3.6 35B A3B, Qwen3.5 122B A10B, and Qwen3 Omni 30B A3B. Reasoning models use extended thinking to work through complex problems before providing an answer.
Yes, 20 of 27 models on Alibaba Cloud are open weight models: Qwen3.8 2.4T A95B, Qwen3.8-Flash-Next, Qwen3.8 27B (xhigh), Qwen3.8 27B (medium), Qwen3.8 27B (low), Qwen3.5 27B, Qwen3.5 397B A17B (non-reasoning), Qwen3.6 27B, Qwen3.8 27B (non-reasoning), Qwen3.6 27B (non-reasoning), Qwen3.5 27B (non-reasoning), Qwen3.5 35B A3B, Qwen3.5 397B A17B, Qwen3.6 35B A3B, Qwen3.5 122B A10B (non-reasoning), Qwen3.5 122B A10B, Qwen3.6 35B A3B (non-reasoning), Qwen3.5 35B A3B (non-reasoning), Qwen3 Omni 30B A3B, and Qwen3 Omni 30B A3B.
Yes, provider performance can vary over time due to infrastructure changes, load balancing, and updates. We continuously benchmark all providers and display historical performance trends in the "Over Time" charts.
When choosing a model on Alibaba Cloud, consider: intelligence (for quality-sensitive tasks), output speed (for throughput-intensive tasks), latency (for interactive applications requiring quick first responses), pricing (for cost-sensitive workloads), and features like context window size, JSON mode, or function calling support.