Qwen3.8 2.4T A95B: Intelligence, Performance and Price Analysis

Qwen3.8 2.4T A95B, released on August 12, 2026, scores 58 on the Artificial Analysis Intelligence Index, well above the median of 27, but is priced at $2.00 per 1M input tokens and $6.00 per 1M output tokens, making it expensive relative to comparable open-weight models. The model, with 2.4 trillion total parameters and 95 billion active parameters, supports a 984k token context window and outputs text at 51 tokens per second, slower than the average of 66.

Qwen3.8 2.4T A95B Intelligence, Performance & Price Analysis Model summary Qwen3.8 2.4T A95B is amongst the leading models in intelligence, but particularly expensive when comparing to other open weight models of similar size. It's also slower than average and somewhat verbose. The model supports text input, outputs text, and has a 984k tokens context window. Qwen3.8 2.4T A95B scores 58 on the Artificial Analysis Intelligence Index, placing it well above average among comparable models median: 27 . When evaluating the Intelligence Index, it generated 140M tokens, which is somewhat verbose in comparison to the median of 100M. Pricing for Qwen3.8 2.4T A95B is $2.00 per 1M input tokens expensive, median: $0.33 and $6.00 per 1M output tokens expensive, median: $1.20 . In total, it cost $1671.89 to evaluate Qwen3.8 2.4T A95B on the Intelligence Index. At 51 tokens per second, Qwen3.8 2.4T A95B is slower than average 66 . | Reasoning | Yes This page shows the reasoning version of this model. A non-reasoning variant may also exist. | |---|---| | Input modality | Supports: text | | Output modality | Supports: text | | Context window | 984k ~1475 A4 pages of size 12 Arial font | | Total parameters | 2400B | | Active parameters | 95B Number of parameters active per token during inference | | License | | Hugging Face https://huggingface.co/Qwen/Qwen3.8-2.4T-A95B Metrics are compared against models of the same class: - Non-reasoning models → compared only with other non-reasoning models - Reasoning models → compared across both reasoning and non-reasoning - Open weights models → compared only with other open weights models of the same size class: - Tiny: ≤4B parameters - Small: 4B–40B parameters - Medium: 40B–150B parameters - Large: 150B parameters - Proprietary models → compared across proprietary and open weights models of the same price range, using a blended 3:1 input/output price ratio: - <$0.15 per 1M tokens - $0.15–$1 per 1M tokens - $1 per 1M tokens Highlights Intelligence Artificial Analysis Intelligence Index Artificial Analysis Intelligence Index by Open Weights / Proprietary Intelligence Evaluations Agentic real-world work tasks, Elo-500 /2000 𝜏³-Banking /evaluations/tau3-banking Updated Agentic tool use Agentic coding & terminal use Coding Humanity's Last Exam /evaluations/humanitys-last-exam Updated Reasoning & knowledge Scientific reasoning Physics reasoning AA-Omniscience Accuracy /evaluations/omniscience Updated Knowledge 1 - hallucination rate AA-LCR /evaluations/artificial-analysis-long-context-reasoning Updated Long context reasoning Agentic knowledge work, Elo Agentic SaaS workflows Legal agentic work, criterion pass rate Agentic business operations Quantitative analysis on spreadsheets & documents Instruction following Long-horizon agentic tasks Kubernetes incident root-cause analysis Visual reasoning AA-Omniscience AA-Omniscience Index Openness Index Artificial Analysis Openness Index: Score Intelligence Index Comparisons Intelligence Index vs. Cost per Intelligence Index Task Token Use Output Tokens per Intelligence Index Task Cost Cost per Intelligence Index Task Cost to Run Artificial Analysis Intelligence Index Pricing: Cache Hit, Input, and Output Context Window Context Window Speed Measured by Output Speed tokens per second Output Speed Time per Intelligence Index Task Latency Measured by Time seconds to First Token Latency: Time To First Answer Token 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 Model Size Open Weights Models Only Model Size: Total and Active Parameters Frequently Asked Questions Common questions about Qwen3.8 2.4T A95B Qwen3.8 2.4T A95B was released on August 12, 2026. Qwen3.8 2.4T A95B was created by Alibaba. Qwen3.8 2.4T A95B scores 58 on the Artificial Analysis Intelligence Index, placing it well above average among other open weight models of similar size median: 27 . Qwen3.8 2.4T A95B generates output at 50.8 tokens per second based on Alibaba's API , which is below average compared to other open weight models of similar size median: 65.6 t/s . Qwen3.8 2.4T A95B has a time to first token TTFT of 2.78s based on Alibaba's API , which is somewhat higher than average compared to other open weight models of similar size median: 1.94s . Qwen3.8 2.4T A95B costs $2.00 per 1M input tokens at the higher end, median: $0.57 and $6.00 per 1M output tokens at the higher end, median: $2.20 , based on Alibaba's API. Qwen3.8 2.4T A95B costs $2.00 per 1M input tokens and $6.00 per 1M output tokens based on Alibaba's API . For a blended rate 7:2:1 cache hit/input/output ratio , this is $1.18 per 1M tokens. Pricing may vary by provider. Compare provider pricing /models/qwen3-8-2-4t-a95b/providers When evaluated on the Intelligence Index, Qwen3.8 2.4T A95B generated 140M output tokens, which is somewhat higher than average compared to other open weight models of similar size median: 100M . Yes, Qwen3.8 2.4T A95B is a reasoning model. It uses extended thinking or chain-of-thought reasoning to work through complex problems before providing an answer. Qwen3.8 2.4T A95B supports text input. Qwen3.8 2.4T A95B supports text output. No, Qwen3.8 2.4T A95B does not support image input. It can only process text. No, Qwen3.8 2.4T A95B is not multimodal. It only supports text input. Qwen3.8 2.4T A95B has a context window of 980k tokens. This determines how much text and conversation history the model can process in a single request. Yes, Qwen3.8 2.4T A95B is open weights. The model weights are publicly available and can be downloaded for self-hosting. Qwen3.8 2.4T A95B has 2.4 trillion parameters 95 billion active . Qwen3.8 2.4T A95B is a Mixture of Experts MoE model with 2.4 trillion total parameters, but only 95 billion active parameters are used during inference. Qwen3.8 2.4T A95B is released under the Qwen3.8-Max License license. Commercial use requires a separate license agreement. View license https://huggingface.co/Qwen/Qwen3.8-2.4T-A95B/blob/main/LICENSE Qwen3.8 2.4T A95B achieves a score of 58 on the Artificial Analysis Intelligence Index. This composite benchmark evaluates models across reasoning, knowledge, mathematics, and coding. Yes, Qwen3.8 2.4T A95B is available via API through 1 provider. Compare API providers /models/qwen3-8-2-4t-a95b/providers Qwen3.8 2.4T A95B is available through 1 API provider. Compare providers /models/qwen3-8-2-4t-a95b/providers