Qwen3.8 2.4T A95B

qwen3.8-2.4t-a95bby QwenZDR (all rungs)

Qwen3.8 2.4T A95B is an open-weight sparse mixture-of-experts model from Qwen and the open-weight variant of [Qwen3.8 Max](/qwen/qwen3.8-max), with 95 billion active parameters out of 2.4 trillion total. It is...

Try in Playground

Context

1.05M

Max output

262K

Price per million tokens

$2in$0.25cached$6out

Released

Aug 2026

Waterfall

1
Fireworks via Experiential Cloud$2/$699.9% up--

Quickstart

I want you to route my LLM calls for "qwen3.8-2.4t-a95b" through the Experiential gateway instead of calling the provider directly. It speaks the OpenAI Chat Completions API, so this is a base-URL and key swap. Please:

1. Point the client at https://api-pr-1072.preview.experientiallabs.ai/v1 as the base URL.
2. Authenticate with my Experiential API key from the EXPLABS_API_KEY environment variable. If it isn't set, stop and tell me to create one under Settings -> API Keys and export it.
3. Use the model id "qwen3.8-2.4t-a95b" exactly.
4. Update every place my code builds an LLM client for this model to use that base URL and key, leaving streaming and tool-calls as they are.
5. Make one test call and show me the reply plus the token usage, so we confirm it runs on my Experiential credits.

Tell me which files you changed.

CapabilitiesAccepts text

Sampling

Temperature0-2Top-p0-1Top-kStop sequences

Tools & structure

ToolsParallel tool callsStructured output

Streaming

Limits

Max output262,144Max-tokens fieldmax_tokens

Sending an unsupported field? See error referenceFireworks docs

Benchmarks2

Benchmark	Score	Source
GPQA Diamondvendor reported · Aug 2026	92.6%	vendor reported · Aug 2026
Terminal-Benchpublic leaderboard · Aug 2026	86.6	public leaderboard · Aug 2026

Providers serving this model1

												Status
Fireworks via Experiential Cloudexperiential	no data	no data	99.9%	#1 / 1	1.05M	262K	$2≈	$6≈	cache$0.25≈	-	ZDR	active	estimated

Hugging Face

Models