Qwen 3.8 vs LongCat-2.0

What I tested — same prompt, two models

I run the same fixed prompt set through every new model the day it drops — same string, one shot, single HTML file out — and I score the result 0–10 on whether it ran, how close it hit the brief, and how good it looked. Below is what came out when I gave the exact same prompts to Qwen 3.8 and LongCat-2.0, side by side, on 3 shared tasks inside the Agent Operating System.

Both models were given identical prompts inside the Agent Operating System — no help, no iteration, no "best of N" tricks. I run each prompt once, save the HTML file the model produces, and score it 0–10 on whether it ran, how close it hit the brief, and how good it looked. The scoring is mine. The verdicts below are pulled from my source comparison guides at agentos.guide where I publish every score and the reasoning behind it.

Qwen 3.8 · Benched on GoldieBench via the Qoder CLI (`qoder-qwen`, model Qwen3.8-Max-Preview) — the only door while it has no public API. Non-game tasks are one-shot like the rest of the field. GAME tasks run in Qoder's real AGENT mode: skill-infused build, then up to 2 QA fix rounds where a vision judge + live console errors are fed back and Qwen 3.8 edits its own file (it took crypt from a black-screen 3.0 to a torch-lit 7.8). Scored on a mid-play frame by the same Opus judge as everyone else.

LongCat-2.0 · Run through the free longcat.chat web chat (the API key had no token quota), driven with the local-model-tester GoldieBench prompts; every build render-verified + playtested (verify-move.js: walks + looks + zero errors) before scoring. Slots into the Agent OS as an open frontier coder via its OpenAI-compatible API or the Claude Code / OpenClaw / Hermes harnesses.

Where Qwen 3.8 beat LongCat-2.0

The tasks where I gave Qwen 3.8 a higher 0–10 score on the same prompt — with the actual commentary from my source guides.

Where LongCat-2.0 beat Qwen 3.8

The tasks where I gave LongCat-2.0 a higher 0–10 score on the same prompt — with the actual commentary from my source guides.

Strengths & weaknesses I logged

Pricing & context — the spec sheet

The verdict — which should you pick?

Across 3 scored shared tasks, the averages are essentially tied — Qwen 3.8 8.07 vs LongCat-2.0 8.33. This isn't the comparison where one wins; it's the comparison where you pick based on context, pricing, and what you're actually trying to ship.

If you only run one of these inside your stack, the head-to-head average above is the call. If you can run both, my honest play is to wire Qwen 3.8 and LongCat-2.0 both into the Agent Operating System and dispatch each from the kanban by task type — one-shot 3d game and world prototypes where atmosphere matters → Qwen 3.8, one-shot single-file 3d / html / game builds inside the agent os → LongCat-2.0. That's the same setup I run for the 4,000+ founders inside the AI Profit Boardroom.

FAQ — Qwen 3.8 vs LongCat-2.0

Which is better, Qwen 3.8 or LongCat-2.0?

On Goldie Bench, Qwen 3.8 averages 8.07/10 across the shared tasks, with 10 gold, 10 silver, 5 bronze overall. LongCat-2.0 averages 8.33/10, with 0 gold, 0 silver, 0 bronze. LongCat-2.0 wins the head-to-head 2–1.

How much does Qwen 3.8 cost vs LongCat-2.0?

Qwen 3.8: Qwen3.8-Max-Preview is Alibaba's ~2.4T-parameter flagship, positioned just behind Claude Fable 5. It is NOT on OpenRouter or a public API yet — the only access today is inside Qoder (Alibaba's agentic coding platform, free 2-week Pro trial). Benched here via the Qoder CLI on model `Qwen3.8-Max-Preview`, one-shot. LongCat-2.0: LongCat-2.0 is open-sourced (weights on Hugging Face + GitHub) and served via the longcat.chat web chat plus an OpenAI-compatible API (model id 'LongCat-2.0' at api.longcat.chat/openai/v1). It's a 1.6T-parameter MoE with ~48B activated per token, trained entirely on AI ASIC superpods (>50K accelerators, 35T+ tokens, no rollbacks). Note: the direct API key we were handed shipped with zero token quota ('Token 额度不足'), so every build here was run through the free web chat. Vendor: Meituan.

What's the context window for Qwen 3.8 vs LongCat-2.0?

Qwen 3.8 has a Served through Alibaba's Qoder agent platform; the 3.8-Max preview has no standalone public context window yet. context window. LongCat-2.0 has a 1,000,000 tokens (LongCat Sparse Attention) context window.

When should I pick Qwen 3.8 over LongCat-2.0?

Pick Qwen 3.8 for: One-shot 3D game and world prototypes where atmosphere matters; Anyone already in the Qoder IDE/CLI wanting a near-frontier model free on the Pro trial; A cheaper stand-in for Fable 5 on creative-visual builds. The trade-off is the weaknesses we logged on the bench: Torch-lit dungeon (crypt) came out generic (6.3); one open-world RPG one-shot black-screened (twilightvale 2.5) — classic three.js r128 API drift; Preview is Qoder / Token-Plan only — no OpenRouter or public API, so it can't be routed into an app the way the OpenRouter models can.

When should I pick LongCat-2.0 over Qwen 3.8?

Pick LongCat-2.0 for: One-shot single-file 3D / HTML / game builds inside the Agent OS; Long-context, repo-level edits + automated agentic task execution; A free, open, frontier-class coder to drop into the Model-Proof System. The trade-off is the weaknesses we logged on the bench: The direct API key we were given had near-zero token quota, so we ran it through the free web chat rather than the API; One camera-framing miss: Voxel Craft loaded facing away from the world (sky-only) until a one-line yaw/pitch patch pointed it at the terrain.

How does Goldie Bench score Qwen 3.8 vs LongCat-2.0?

Every demo on this page was built by Julian Goldie inside the Agent Operating System — same fixed prompt for both models, one shot, single HTML file out. Each result gets a 0–10 score on whether it ran, how close it hit the brief, and how good it looked. The highest score on each task gets gold; second gets silver; third gets bronze. See methodology for full provenance.

Related comparisons

Other head-to-heads using the same scoring system:

Full model pages: Qwen 3.8 · LongCat-2.0 · back to the leaderboard