Fireworks AI model
Qwen3.8-2.4T-A95B
Qwen3.8-2.4T-A95B is Alibaba's most capable Qwen model to date, a 2.4T-parameter sparse MoE with ~95B active parameters. It is built for autonomous, long-horizon work: multi-day coding runs, research-paper reproduction and self-improvement.
- Provider
- Fireworks AI
- Availability
- Available
- Model ID
- qwen3p8-2p4t-a95b
- Snapshot
- 2026-10-07
Qwen3.8-2.4T-A95B Benchmarks & Agentic Accuracy
Verified primary research & developer preference metrics
What This Means in Real Software Development
- Massive 2.4T total parameter capacity with 95B active parameters per token.
- 64.0% SWE-bench Verified pass rate with advanced chain-of-thought code reasoning.
- High inference throughput (~145 TPS) on modern inference clusters.
- Priced competitively ($0.25/M input, $0.75/M output) on Fireworks AI.
Swarm Multi-Agent Topology
Best Paired With: Claude 3.7 Sonnet as Lead Planner
High throughput and strong reasoning make it ideal for parallel Coder subagents in Swarm.
Dispatched across isolated Git worktrees to parallelize bug fixing and test suites.
What Developers & Practitioners Are Saying
Real-world field observations, production strengths, and developer consensus
"Massive mixture-of-experts flagship from Alibaba Cloud, delivering near-frontier performance across enterprise benchmarks."
Verified Real-World Strengths
- 64.0% SWE-bench Verified
- Huge 2.4T parameter pretraining base providing vast domain breadth
- Exceptional performance on complex database queries, DevOps, and shell scripting
Production Caveats & Gaps
- Requires enterprise multi-GPU hardware (H100/H200 cluster) to host locally
- API throughput varies depending on provider infrastructure
Primary Sources & Methodology Citations
- Qwen 3.8 Architecture and MoE Benchmarks — Alibaba Cloud Qwen Team (February 2026)
Qwen3.8-2.4T-A95B model overview
Qwen3.8-2.4T-A95B is listed in the public Swarm catalog through Fireworks AI. This page summarizes the verified catalog fields used by Swarm without claiming real-time access for every provider account.
- Provider
- Fireworks AI
- Family
- Qwen
- Context window
- 262.1K tokens
- Thinking support
- off · low · medium · high
- Availability status
- Available
- Catalog ID
- fireworks/qwen3p8-2p4t-a95b
Run Qwen3.8-2.4T-A95B in Swarm
Verified CLI workflow & multi-agent architecture
swarm ctl model set --provider fireworks --model fireworks/qwen3p8-2p4t-a95b# Or switch interactively inside the terminal TUI:
/modelsSubagent Role Matrix
Assign Qwen3.8-2.4T-A95B to specialized system roles in ~/.swarm/config.json:
- Primary / Planner: Orchestrates full task DAGs and multi-stage pipelines.
- Coder Subagent: Generates code and unit tests inside private Git worktrees.
- Finder Subagent: Performs read-only code search, symbol discovery, and mapping.
Public catalog snapshot. Catalog availability is not a real-time provider status or an access guarantee.