Qwen3.8-2.4T-A95B
Official · BF16 · safetensors
Runtime certified: Transformers vLLM SGLang
| Precision / quantization | Format | Bits | Publisher trust | Deployment state |
|---|---|---|---|---|
| BF16 | safetensors | 16 | Official | Runtime certified |
Qwen · General · Qwen3.8
Qwen3.8 2.4T-A95B: Specs, GPU requirements, hosting & deployment
Qwen3.8 2.4T-A95B is a current General release in the Qwen family, certified from its immutable official source revision.
Total parameters2,400.0B
Active parameters95.0B
Context window262,144 tokens
Release statusCurrent · Current
ArchitectureQwen3_5MoeForCausalLM
ModalitiesText
LicenseQwen3.8-Max License
Deployment stateDeployable with certified configuration
Commercial stateCommercial Use Allowed With Conditions
Release date2026-08-08
Architecture, strengths and use cases
Qwen3.8 2.4T-A95B is an official self-hostable Qwen release in the Qwen General track. It uses Qwen3_5MoeForCausalLM, carries 2400B / 95B MoE, supports a verified context window of 262,144 tokens, and exposes only source-certified checkpoints and runtime combinations. Its release, license and deployment facts are tied to the official repository revision rather than inferred from the family name.
Qwen3.8 2.4T-A95B uses Qwen3_5MoeForCausalLM with 2400B / 95B MoE. The stored architecture profile records 92 layers, hidden width 8192, 64 attention heads and a 262,144 token context ceiling where the official configuration publishes those fields.
Verified Qwen3_5MoeForCausalLM architecture
262,144 token certified context
Official BF16, FP8, Unknown checkpoint coverage
Private LLM inference
Agentic workflows
Long-context document processing
Certified checkpoints and quantizations
Only exact, source-certified variants attached to this release are shown.
Qwen3.8-2.4T-A95B-FP8
Official · FP8 · safetensors
Runtime certified: Transformers vLLM SGLang
| Precision / quantization | Format | Bits | Publisher trust | Deployment state |
|---|---|---|---|---|
| FP8 | safetensors | 8 | Official | Runtime certified |
Recommended hardware
Use only the listed transformers, vllm, sglang runtime configuration with the exact checkpoint precision. The recommendation engine checks weight memory, KV cache, runtime overhead, safety headroom, GPU architecture and topology before ranking rentable and reference hardware.
Estimated weight memory4604.46 GB
KV cache2.88 GB
Runtime overhead885.96 GB
Total with safety margin6811.68 GB
Technically compatible reference sizing
No current FusionPointAI inventory passes every runtime, architecture, topology and memory gate for this exact checkpoint. Use the estimator for global compatible hardware.
License and self-hosting considerations
The full 2400B resident weight set must fit across the selected GPUs. Context and concurrency add KV-cache memory; MoE active parameters do not replace resident-weight memory.
Qwen3.8 2.4T-A95B is published under Qwen3.8-Max License. The release-scoped official license link and verification date are retained with the catalogue record.
Commercial Use Allowed With Conditions. Commercial use is conditional. Review the linked official terms before deployment.
Alternatives and sibling variants
Qwen3.8 27B, Qwen3 30B-A3B Instruct 2507, Qwen3 235B-A22B
Curated comparisons
Release timeline
qwen/qwen/qwen3-8-2-4t-a95b @ 74e316d144b8d12c4c435d77676e436595e7235267d84610e0c3d17b49399151
Frequently asked questions
How much VRAM does Qwen3.8 2.4T-A95B require?
The exact estimate depends on checkpoint precision, context and concurrency; the page recommendation uses the certified memory engine and compatible runtime configurations.
Which runtimes support Qwen3.8 2.4T-A95B?
Certified upstream runtime evidence is available for transformers, vllm, sglang.
Can Qwen3.8 2.4T-A95B be used commercially?
The stored commercial classification is Commercial Use Allowed With Conditions. Consult the official Qwen3.8-Max License terms for the final obligations.
Official references
Last verified: 2026-08-26