
Real head-to-head · same prompt, one shot
Claude Opus 5.5 vs Hy3
Anthropic's Opus 5.5 — benched on all 50 one-shot builds, every game playtested. vs Tencent's open-weights coder — Apache-2.0, cheap, beats GLM-5.1 on frontend in Tencent's blind eval.
Head-to-head verdict: Claude Opus 5.5 wins 6–1.
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 Claude Opus 5.5 and Hy3, side by side, on 7 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.
Claude Opus 5.5 · Benched on all 50 GoldieBench tasks via a Claude subscription: one-shot builds, real renders, pixel-diff playtests on every game, then vision-judged.
Hy3 · Wired into the Agent OS as the 'Hy3 Coder' tab (chat + live preview + workspace) via OpenRouter. Bench built one-shot on the same prompts as the field; weak builds iterated by Hy3 itself (the model fixes its own builds).
Side-by-side on 50 shared tasks
Click any cell to play that model's actual one-shot attempt. Medals are derived from my 0–10 scores per task (highest = 🥇, second = 🥈, third = 🥉).
Task ↓
Claude Opus 5.5
Hy3
Game
Game
Game
Game
Game
Game
Page
Game
— not attempted —
Game
— not attempted —
Game
— not attempted —
Game
— not attempted —
Game
— not attempted —
Game
— not attempted —
Game
— not attempted —
Game
— not attempted —
Game
— not attempted —
Game
— not attempted —
Game
— not attempted —
Game
— not attempted —
Game
— not attempted —
Game
— not attempted —
Game
— not attempted —
Game
— not attempted —
Game
— not attempted —
Where Claude Opus 5.5 beat Hy3
The tasks where I gave Claude Opus 5.5 a higher 0–10 score on the same prompt — with the actual commentary from my source guides.
Flightsim
Game
Claude Opus 5.5 8.2
·
Hy3 6.8
(+1.4)
What I saw: Covers the whole brief: a real takeoff roll, heading tape, attitude ball, airspeed/altitude/VS readouts, gear, graded landings and crash states. The airfield has a hangar and tower and the prop plane is detailed. The frame is still on the runway and the sun glare blows out the to…
Doom
Game
Claude Opus 5.5 5.8
·
Hy3 4.5
(+1.3)
What I saw: Solid raycaster chrome (shotgun viewmodel, minimap with demons, cohesive HUD), but the mid-play frame is a red-washed, noisy wall maze with no demon on screen. The known splice bug breaks the KILLS/DEMONS LEFT count and the win check after the first kill.
Dragonrealm
Game
Claude Opus 5.5 8.3
·
Hy3 7.2
(+1.1)
What I saw: Charming frozen world: snowfall, moon, snow-capped pines, ruins arch, a modelled armoured hero with a drawn sword, a wraith enemy, and a cohesive HP/ST/compass/dragon HUD. Low-poly and stylised rather than epic, and the fog-flattened snow reads a little bland, so it falls short o…
Parachute
Game
Claude Opus 5.5 7.6
·
Hy3 6.8
(+0.8)
What I saw: The exit moment works: a detailed plane, an articulated skydiver, altitude/descent HUD, distance-to-clearing compass and a deploy prompt. The mid-play frame is mostly empty grey-blue sky, and no jungle or clearing is visible yet.
Gtadrive
Game
Claude Opus 5.5 8.0
·
Hy3 7.4
(+0.6)
What I saw: All the open-city parts are in place: windowed tower canyons under a sunset sky, cop cars already chasing, wanted stars, cash goal, minimap and steal-car key. A huge foreground streetlamp and blotchy pink light cones clutter the view, and the gun turret on the car is a generic touch.
Where Hy3 beat Claude Opus 5.5
The tasks where I gave Hy3 a higher 0–10 score on the same prompt — with the actual commentary from my source guides.
Aipbpromo
Page
Hy3 7.4
·
Claude Opus 5.5 7.2
(+0.2)
What I saw: Clean multi-scene pipeline with animated count-up stats, gold/cyan cinematic palette, particle field and progress bar render correctly; but the stats appear left-clustered and off-center with only two of four visible mid-animation, feeling sparse rather than composed, keeping it …
Strengths & weaknesses I logged
Claude Opus 5.5
Strengths
- Frontier-class coding + agentic reasoning (Claude 5 family)
- 1M-token context — reads an entire codebase in one call
- All 50 one-shot builds rendered with zero console errors; all 23 games passed the input playtest
Trade-offs
- Premium pricing ($4/$20 per M) — route the everyday 90% to cheaper lanes
- One-shot builds can still ship logic bugs (e.g. a doom kill counter that miscounts)
Hy3
Strengths
- Apache-2.0 open weights — self-host free, no lock-in
- Tencent's 270-expert blind eval: 2.67/4 vs GLM-5.1's 2.51, strongest on frontend / data / CI-CD
- Hallucination rate cut 12.5% → 5.4%; stable tool-calls across scaffoldings (<4% SWE-Bench variance)
Trade-offs
- Slow upstream on OpenRouter (30-90s per build) — fine for one-shots, sluggish for tight loops
- One-shot game builds can under-render (flat raycaster walls, unlit 3D) without an iterate pass
Pricing & context — the spec sheet
| Spec | Claude Opus 5.5 | Hy3 |
|---|---|---|
| Vendor | Anthropic | Tencent Hunyuan |
| Context window | 1,000,000 tokens | 262,144-token context window. Open weights (Apache-2.0) on HuggingFace / ModelScope / GitHub; benched here via OpenRouter. |
| Price | $4 / $20 per M | $0.14 / 1M input · $0.58 / 1M output |
| Pricing detail | Anthropic's Opus 5.5 — 1M-token context, listed at $4 input / $20 output per million tokens. Benched through a Claude subscription; game tasks use our skill-infused AAA build prompts. | Tencent Hunyuan 3 — open-weights under Apache-2.0, so free to self-host. On OpenRouter it is one of the cheapest capable coders: ~$0.14/M in, $0.58/M out (1 RMB / 4 RMB). Upstream can be slow (30-90s to first token), but per-token cost is negligible. |
| Release | 2026-09 | 2026-07-06 |
| Bench coverage | 50/50 scored · avg 7.57/10 | 7/7 scored · avg 6.76/10 |
The verdict — which should you pick?
Across 7 scored shared tasks, Claude Opus 5.5 averaged 7.53/10, beating Hy3's 6.76/10 by 0.77 points. Pick Claude Opus 5.5 when the build has to ship on the first prompt and you can afford the trade-offs in the comparison below.
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 Claude Opus 5.5 and Hy3 both into the Agent Operating System and dispatch each from the kanban by task type — hardest agentic builds → Claude Opus 5.5, cost-sensitive coding + frontend design where open weights matter → Hy3. That's the same setup I run for the 4,000+ founders inside the AI Profit Boardroom.
FAQ — Claude Opus 5.5 vs Hy3
Which is better, Claude Opus 5.5 or Hy3?
On Goldie Bench, Claude Opus 5.5 averages 7.53/10 across the shared tasks, with 0 gold, 0 silver, 1 bronze overall. Hy3 averages 6.76/10, with 0 gold, 0 silver, 0 bronze. Claude Opus 5.5 wins the head-to-head 6–1.
How much does Claude Opus 5.5 cost vs Hy3?
Claude Opus 5.5: Anthropic's Opus 5.5 — 1M-token context, listed at $4 input / $20 output per million tokens. Benched through a Claude subscription; game tasks use our skill-infused AAA build prompts. Hy3: Tencent Hunyuan 3 — open-weights under Apache-2.0, so free to self-host. On OpenRouter it is one of the cheapest capable coders: ~$0.14/M in, $0.58/M out (1 RMB / 4 RMB). Upstream can be slow (30-90s to first token), but per-token cost is negligible.
What's the context window for Claude Opus 5.5 vs Hy3?
Claude Opus 5.5 has a 1,000,000 tokens context window. Hy3 has a 262,144-token context window. Open weights (Apache-2.0) on HuggingFace / ModelScope / GitHub; benched here via OpenRouter. context window.
When should I pick Claude Opus 5.5 over Hy3?
Pick Claude Opus 5.5 for: Hardest agentic builds; Whole-repo reasoning; Frontier one-shots. The trade-off is the weaknesses we logged on the bench: Premium pricing ($4/$20 per M) — route the everyday 90% to cheaper lanes; One-shot builds can still ship logic bugs (e.g. a doom kill counter that miscounts).
When should I pick Hy3 over Claude Opus 5.5?
Pick Hy3 for: Cost-sensitive coding + frontend design where open weights matter; Self-hosters who want an Apache-2.0 model they fully own; Anyone wiring a cheap capable coder into a live build panel (Agent OS Hy3 Coder tab). The trade-off is the weaknesses we logged on the bench: Slow upstream on OpenRouter (30-90s per build) — fine for one-shots, sluggish for tight loops; One-shot game builds can under-render (flat raycaster walls, unlit 3D) without an iterate pass.
How does Goldie Bench score Claude Opus 5.5 vs Hy3?
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:
Claude Opus 5.5 vs Fusion Hy3 vs Fusion Claude Opus 5.5 vs Claude Opus 5 Hy3 vs Claude Opus 5 Claude Opus 5.5 vs Hermes MoA Hy3 vs Hermes MoA Claude Opus 5.5 vs GPT-5.6 Sol Hy3 vs GPT-5.6 SolFull model pages: Claude Opus 5.5 · Hy3 · back to the leaderboard
The same stack Julian uses
Run this stack yourself.
Every demo on this bench was built inside the Agent Operating System — one prompt, one shot, single HTML file out. The Agent OS, the prompts, the templates, the weekly walkthroughs and 4,000+ founders shipping with it every day all live inside the AI Profit Boardroom.
4,000+founders
258documented wins
38countries
$59/momonthly





























