Qwen3.8 2.4T A95B
Summary
Qwen3.8 2.4T A95B is an open-weight language model from Alibaba (Qwen), released on 12 Aug 2026. It accepts text and images and generates text, with a 262K-token context window and up to 131K output tokens per response. As of 18 Nov 2026, the lowest listed API price is $2 per million input tokens and $6 per million output tokens (Fireworks AI), across 6 providers we track.
Open-weight sparse MoE (2.4T total, 95B active), the open-weight twin of Qwen3.8 Max for coding, research, complex reasoning, and agentic workflows
Last verified 18 Nov 2026. Every figure on this page links to its source.
Specifications
| Developer | Alibaba (Qwen) |
|---|---|
| Release date | 12 Aug 2026 [models.dev] |
| Status | Generally available [models.dev] |
| Weights | Open weights [Hugging Face] |
| License | qwen3.8-max [Hugging Face] |
| Parameters | 2.4T [Hugging Face] |
| Context window | 262K (262,144 tokens) [models.dev] |
| Max output | 131K tokens [models.dev] |
| Input | Text and images [models.dev] |
| Output | Text [models.dev] |
| Reasoning mode | Yes [models.dev] |
| Tool calling | Yes [models.dev] |
| Structured output | Yes [models.dev] |
| Hugging Face | Qwen/Qwen3.8-2.4T-A95B [Hugging Face] |
Qwen3.8 2.4T A95B API pricing by provider
USD per million tokens. Sorted by input price.
| Provider | Input | Output | Cached input | Context | Source |
|---|---|---|---|---|---|
| Fireworks AI accounts/fireworks/models/qwen3p8-2p4t-a95b | $2 | $6 | $0.25 | 262K | models.dev, 18 Nov 2026 |
| Novita AI qwen/qwen3.8-2.4t-a95b | $2 | $6 | $0.25 | 1M | models.dev, 18 Nov 2026 |
| SiliconFlow Qwen/Qwen3.8-2.4T-A95B | $2 | $6 | $0.25 | 1M | models.dev, 18 Nov 2026 |
| OpenRouter qwen/qwen3.8-2.4t-a95b | $2 | $6 | $0.25 | 1M | OpenRouter, 18 Nov 2026 |
| Vercel AI Gateway alibaba/qwen3.8-2.4t-a95b | $2 | $6 | $0.25 | 262K | models.dev, 18 Nov 2026 |
| Hugging Face Inference Providers Qwen/Qwen3.8-2.4T-A95B | $2.50 | $6.25 | – | 262K | models.dev, 18 Nov 2026 |
Running Qwen3.8 2.4T A95B locally
Estimated memory, from the parameter count and the architecture in config.json on Hugging Face. Includes about 0.8 GB of runtime overhead.
| Quantization | Weights | Total, 8K context | Total, 32K context |
|---|---|---|---|
| Q4_K_M (4-bit) | 1483 GB | 1485 GB | 1487 GB |
| Q8_0 (8-bit) | 2599 GB | 2601 GB | 2603 GB |
| FP16 / BF16 (16-bit) | 4892 GB | 4894 GB | 4896 GB |
At 4-bit with an 8K conversation, Qwen3.8 2.4T A95B needs about 1485 GB, more than any single consumer graphics card; it needs a unified-memory machine, a multi-GPU setup or a split between graphics card and system RAM. Check your own hardware, or see Best computer for local AI on our sister site AIPCs.
Frequently asked questions
What is the context window of Qwen3.8 2.4T A95B?
Qwen3.8 2.4T A95B has a context window of 262,144 tokens (262K) and can generate up to 131,072 tokens in a single response, according to models.dev.
How much does the Qwen3.8 2.4T A95B API cost?
Qwen3.8 2.4T A95B is listed from $2 per million input tokens and $6 per million output tokens (Fireworks AI). Prices last verified 18 Nov 2026.
Is Qwen3.8 2.4T A95B open source?
Qwen3.8 2.4T A95B is an open-weight model: its weights are published on Hugging Face as Qwen/Qwen3.8-2.4T-A95B under the qwen3.8-max license. Check the license terms before commercial use.
When was Qwen3.8 2.4T A95B released?
Alibaba (Qwen) released Qwen3.8 2.4T A95B on 12 Aug 2026, according to models.dev.
Which providers offer Qwen3.8 2.4T A95B?
We track Qwen3.8 2.4T A95B on 6 providers: Fireworks AI, Novita AI, SiliconFlow, OpenRouter, Vercel AI Gateway, Hugging Face Inference Providers.
How much VRAM does Qwen3.8 2.4T A95B need?
About 1485 GB at 4-bit (Q4_K_M) with an 8K-token conversation, and about 2601 GB at 8-bit. That is more than any single consumer graphics card holds. These are estimates from the model's published parameter count and architecture.
Similar models from other labs
Comparable on output type, weights, release date, price and context window. Prices are the lowest listed, USD per million tokens.
| Model | Lab | Released | Context | Input $/1M | Output $/1M | Weights |
|---|---|---|---|---|---|---|
| Aion 3.5 | Aion Labs | 23 Sept 2026 | 262K | $3 | $6 | – |
| Kimi K2.7 Code High Speed | Moonshot AI | 12 Jun 2026 | 262K | $1.90 | $8 | Open |
| Pareto 26.9 | Unbiased | 13 Oct 2026 | 262K | $2.50 | $7.50 | – |
| Inkling | Thinkingmachines | 15 Jul 2026 | 256K | $1 | $4.05 | Open |
| Mistral Medium Latest | Mistral AI | 30 Apr 2026 | 262K | $1.50 | $7.50 | Open |
| Mistral Medium 3.5 | Mistral AI | 29 Apr 2026 | 262K | $1.50 | $7.50 | Open |
Change history
- Input modalities changed from text to text, image
- Context window changed from 1M to 262K
- Max output changed from 131K to 131K
- Context window changed from 1M to 1M
- Context window changed from 1M to 1M
- Max output changed from 131K to 131K
- Context window changed from 262K to 1M
- Input modalities changed from text, image to text
- Max output changed from 128K to 131K
- Context window changed from 1M to 262K
- Input modalities changed from text to text, image
- Max output changed from 131K to 128K
- Input modalities changed from text, image to text
- Max output changed from 128K to 131K
- Context window changed from 262K to 1M
- Context window changed from 1M to 262K
- Input modalities changed from text to text, image
- Max output changed from 131K to 128K
- Context window changed from 1M to 1M
- Max output changed from 131K to 131K
- Context window changed from 262K to 1M
- Context window changed from 1M to 262K
- Max output changed from 131K to 131K
- Context window changed from 1M to 1M
- Input modalities changed from text, image to text
Sources
- https://openrouter.ai/qwen/qwen3.8-2.4t-a95b
- https://github.com/sst/models.dev/blob/dev/providers/vercel/models/alibaba/qwen3.8-2.4t-a95b.toml
- https://huggingface.co/Qwen/Qwen3.8-2.4T-A95B
- https://github.com/sst/models.dev/blob/dev/providers/fireworks-ai/models/accounts/fireworks/models/qwen3p8-2p4t-a95b.toml
- https://github.com/sst/models.dev/blob/dev/providers/novita-ai/models/qwen/qwen3.8-2.4t-a95b.toml
- https://github.com/sst/models.dev/blob/dev/providers/siliconflow/models/Qwen/Qwen3.8-2.4T-A95B.toml
- https://github.com/sst/models.dev/blob/dev/providers/huggingface/models/Qwen/Qwen3.8-2.4T-A95B.toml