⭐ Get the Agent OS + join 4,000+ founders inside the AI Profit Boardroom → Join AIPB ($59/mo)
MiniMax

MiniMax M3

1M-context frontier model at $0.30/M tokens — cheapest big-context model on the bench.

Context1,048,576-token context — matches GLM-5.2 and Fable 5
Pricing$0.30 / 1M input tokens, $1.50 / 1M output
Tasks tested47
Avg score7.97/10 average
Medals🥇2 🥈1 🥉4
Release2026-06-18
Official siteminimax.io ↗
Official vendor source
MiniMax M3 is built by MiniMax — see the vendor's own product page, pricing, and docs at minimax.io.
Visit minimax.io →

Reference benchmarks for MiniMax M3

These are external benchmarks I pulled from the source comparison guides on agentos.guide — SWE-bench Verified, DRACO, Kilo plan rubric, build-time measurements, vendor-reported coding scores. They are not goldiebench medal scores (those come only from same-prompt one-shot creative coding tasks in the matrix). I surface them here so the spec sheet for MiniMax M3 is honest about what's measured.

Context window
1,048,576 tokens
source: /openrouter
Per-token cost
$0.30 / M input · $1.50 / M output
source: /openrouter

What is MiniMax M3?

MiniMax M3 is the MiniMax frontier model with a 1,048,576-token context — matches GLM-5.2 and Fable 5 context window, released 2026-06-18. Tagline: 1M-context frontier model at $0.30/M tokens — cheapest big-context model on the bench.. Official source: minimax.io.

Pricing detail. MiniMax M3 is the cheapest 1M-context frontier model on the bench — roughly 1/200th the per-call cost of OpenRouter Fusion and 1/30th of Claude Opus 4.8. Designed for high-volume agent workloads where context length matters but per-call budget is tight.

How I use it inside the Agent OS. Bench prompts dispatched via OpenRouter. Scored by Claude judge against the same 42 prompts every other model ran.

What I built with MiniMax M3

Every model on Goldie Bench gets the same fixed prompt set — one shot, single HTML file out — and I score the result 0–10 inside the Agent Operating System. Here's what MiniMax M3 shipped on the bench: 47 one-shot demos across 1,048,576-token context — matches GLM-5.2 and Fable 5 of context. Of those, 47 are scored against the field with my honest 0–10 from the source guides at agentos.guide.

Strengths

  • 1M token context — full repo / full deep-research corpus fits in one call
  • $0.30/M input is roughly 1/30th of Opus 4.8 — built for high-volume agent loops
  • Solid one-shot HTML output — clean structure on game and visual prompts

Trade-offs

  • Less polished than Fusion's panel-ensembled output on the toughest deep builds
  • Newer model — less community calibration vs Fable 5 / Opus 4.8

Best for

  • High-volume agent workflows where per-call cost dominates
  • 1M-context tasks (whole-repo refactors, deep-research synthesis)
  • Drop-in cheaper alternative to GLM-5.2 with comparable 1M context

Every benchmark — MiniMax M3's full scorecard

All 47 scored tasks, best first — the judge's 0–10 on the same rubric as the whole field. Click any bar for that task's cross-model page, or open this scorecard in the interactive graphs. Full editorial breakdown with judge quotes and sourced outside research: the MiniMax M3 deep dive →.

DragonrealmGame9.0 🥇RacingGame9.0 🥇TwilightvaleGame9.0 🥉NordiccryptGame9.0 🥈CryptGame8.5 🥉SkyrimGame8.5 VoxelcraftGame8.5 DogfightGame8.5 DragonflightGame8.5 🥉NeonblasterGame8.5 WebosPage8.5 FractalSim8.5 OrbitSim8.5 PathtracerSim8.5 ReactiondiffSim8.5 🥉DoomGame8.0 GtafootGame8.0 GtadriveGame8.0 AipbpromoPage8.0 ParachuteGame8.0 FlightsimGame8.0 ArcadeGame8.0 GameGame8.0 NeonracerGame8.0 RpgGame8.0 LandingPage8.0 ClothSim8.0 ParticleforgeSim8.0 MatrixVisual8.0 VoxelVisual8.0 NeoncityGame7.5 OutrunGame7.5 PoolGame7.5 BoidsSim7.5 FluidSim7.5 GalaxySim7.5 SolarSim7.5 WormholeSim7.5 AuroraVisual7.5 FireworksVisual7.5 LavalampVisual7.5 SynthwaveVisual7.5 TerrainVisual7.5 WavesVisual7.5 RaycasterGame7.0 BlackholeSim7.0 PlasmaVisual6.0

Every demo by MiniMax M3

47 live demos, sorted by category. Click any tile to play the actual one-shot result. Verdicts and 0–10 scores are pulled from the source guides where I posted them publicly.

MiniMax M3 one-shot build of Crypt — GoldieBench AI benchmark screenshot▶ LIVE
Crypt 🥉
Game
Nordic dungeon crawler on three.js — torch-lit corridors, skeletons.
MiniMax M3 one-shot build of Dragonrealm — GoldieBench AI benchmark screenshot▶ LIVE
Dragonrealm 🥇
Game
34KB frozen open world — snowy mountains, pines, flying dragon, full HUD.
MiniMax M3 one-shot build of Doom — GoldieBench AI benchmark screenshot▶ LIVE
Doom
Game
Raycaster + sprite enemies + gun + HUD. 21KB of game logic.
MiniMax M3 one-shot build of Racing — GoldieBench AI benchmark screenshot▶ LIVE
Racing 🥇
Game
59KB third-person arcade racer. Banking turns, speed boost, drift, lap timer.
MiniMax M3 one-shot build of Skyrim — GoldieBench AI benchmark screenshot▶ LIVE
Skyrim
Game
43KB Skyrim attempt on three.js — snowy terrain, pines, dragon, HUD.
MiniMax M3 one-shot build of Twilightvale — GoldieBench AI benchmark screenshot▶ LIVE
Twilightvale 🥉
Game
47KB — densest open-world. Village, NPCs, combat, day/night, weather, inventory.
MiniMax M3 one-shot build of Voxelcraft — GoldieBench AI benchmark screenshot▶ LIVE
Voxelcraft
Game
27KB Minecraft-style sandbox — break/place blocks, hotbar, day/night cycle.
MiniMax M3 one-shot build of Gtafoot — GoldieBench AI benchmark screenshot▶ LIVE
Gtafoot
Game
30KB · plays clean · three, webgl (re-rolled)
MiniMax M3 one-shot build of Gtadrive — GoldieBench AI benchmark screenshot▶ LIVE
Gtadrive
Game
33KB · plays clean · three, webgl
MiniMax M3 one-shot build of Aipbpromo — GoldieBench AI benchmark screenshot▶ LIVE
Aipbpromo
Page
95KB · plays clean · rAF
MiniMax M3 one-shot build of Parachute — GoldieBench AI benchmark screenshot▶ LIVE
Parachute
Game
26KB · plays clean · three, webgl
MiniMax M3 one-shot build of Flightsim — GoldieBench AI benchmark screenshot▶ LIVE
Flightsim
Game
28KB · plays clean · three, webgl
MiniMax M3 one-shot build of Arcade — GoldieBench AI benchmark screenshot▶ LIVE
Arcade
Game
Neon Breakout — paddle, ball, brick wall, particle trails, score HUD.
MiniMax M3 one-shot build of Dogfight — GoldieBench AI benchmark screenshot▶ LIVE
Dogfight
Game
25KB 3D dogfight with enemy AI, missiles, guns.
MiniMax M3 one-shot build of Dragonflight — GoldieBench AI benchmark screenshot▶ LIVE
Dragonflight 🥉
Game
Fly a dragon through neon rings — full HUD, score, fire-breath gauge.
MiniMax M3 one-shot build of Game — GoldieBench AI benchmark screenshot▶ LIVE
Game
Game
M3 chose an arcade build with HUD + score + lives + sound effects.
MiniMax M3 one-shot build of Neonblaster — GoldieBench AI benchmark screenshot▶ LIVE
Neonblaster
Game
30KB neon space shooter — waves, bosses, power-ups, screen-shake.
MiniMax M3 one-shot build of Neoncity — GoldieBench AI benchmark screenshot▶ LIVE
Neoncity
Game
Cyberpunk flythrough with neon towers + light trails.
MiniMax M3 one-shot build of Neonracer — GoldieBench AI benchmark screenshot▶ LIVE
Neonracer
Game
Top-down neon racer with vapor trails + drift physics + lap timer.
MiniMax M3 one-shot build of Nordiccrypt — GoldieBench AI benchmark screenshot▶ LIVE
Nordiccrypt 🥈
Game
41KB Nordic crypt with torch-lit corridors, chasing skeletons, boss room.
MiniMax M3 one-shot build of Outrun — GoldieBench AI benchmark screenshot▶ LIVE
Outrun
Game
Pseudo-3D OutRun racer.
MiniMax M3 one-shot build of Pool — GoldieBench AI benchmark screenshot▶ LIVE
Pool
Game
Canvas-2D billiards with 16 balls, pockets, click-drag aim.
MiniMax M3 one-shot build of Raycaster — GoldieBench AI benchmark screenshot▶ LIVE
Raycaster
Game
Canvas-2D raycaster — WASD walking, textured walls, distance fog.
MiniMax M3 one-shot build of Rpg — GoldieBench AI benchmark screenshot▶ LIVE
Rpg
Game
29KB top-down RPG with tilemap, NPCs, combat, inventory.
MiniMax M3 one-shot build of Landing — GoldieBench AI benchmark screenshot▶ LIVE
Landing
Page
Nova-1 landing with animated gradient hero, three feature cards, footer CTA.
MiniMax M3 one-shot build of Webos — GoldieBench AI benchmark screenshot▶ LIVE
Webos
Page
38KB working desktop — wallpaper, dock, draggable Notes/Paint/Terminal/Calculator windows.
MiniMax M3 one-shot build of Blackhole — GoldieBench AI benchmark screenshot▶ LIVE
Blackhole
Sim
Minimal gravitational-lens shader.
MiniMax M3 one-shot build of Boids — GoldieBench AI benchmark screenshot▶ LIVE
Boids
Sim
Boids flocking with separation/alignment/cohesion + mouse repulsion.
MiniMax M3 one-shot build of Cloth — GoldieBench AI benchmark screenshot▶ LIVE
Cloth
Sim
Verlet cloth sim, draggable, pinnable corners. 17KB clean implementation.
MiniMax M3 one-shot build of Fluid — GoldieBench AI benchmark screenshot▶ LIVE
Fluid
Sim
2D fluid sim with click-drag injection.
MiniMax M3 one-shot build of Fractal — GoldieBench AI benchmark screenshot▶ LIVE
Fractal
Sim
WebGL Mandelbrot shader with click-to-zoom, hold-to-continuous-zoom.
MiniMax M3 one-shot build of Galaxy — GoldieBench AI benchmark screenshot▶ LIVE
Galaxy
Sim
Spiral galaxy on three.js — particle stars, slow rotation.
MiniMax M3 one-shot build of Orbit — GoldieBench AI benchmark screenshot▶ LIVE
Orbit
Sim
44KB top-down orbit map — Mercury through Mars with accurate relative speeds, hover info cards.
MiniMax M3 one-shot build of Particleforge — GoldieBench AI benchmark screenshot▶ LIVE
Particleforge
Sim
Particle sculptor with mouse gravity + preset modes + FPS counter.
MiniMax M3 one-shot build of Pathtracer — GoldieBench AI benchmark screenshot▶ LIVE
Pathtracer
Sim
62KB WebGL shader path tracer with sample accumulation.
MiniMax M3 one-shot build of Reactiondiff — GoldieBench AI benchmark screenshot▶ LIVE
Reactiondiff 🥉
Sim
31KB Gray-Scott shader with click-to-seed.
MiniMax M3 one-shot build of Solar — GoldieBench AI benchmark screenshot▶ LIVE
Solar
Sim
M3's solar — three.js scene with sun + planets + orbits.
MiniMax M3 one-shot build of Wormhole — GoldieBench AI benchmark screenshot▶ LIVE
Wormhole
Sim
3D tunnel flythrough with distorted starfield.
MiniMax M3 one-shot build of Aurora — GoldieBench AI benchmark screenshot▶ LIVE
Aurora
Visual
Aurora ribbons over mountain silhouette.
MiniMax M3 one-shot build of Fireworks — GoldieBench AI benchmark screenshot▶ LIVE
Fireworks
Visual
Click-to-launch fireworks with particle trails.
MiniMax M3 one-shot build of Lavalamp — GoldieBench AI benchmark screenshot▶ LIVE
Lavalamp
Visual
Metaball lava-lamp shader with warm gradient.
MiniMax M3 one-shot build of Matrix — GoldieBench AI benchmark screenshot▶ LIVE
Matrix
Visual
Classic Matrix rain — falling green glyphs.
MiniMax M3 one-shot build of Plasma — GoldieBench AI benchmark screenshot▶ LIVE
Plasma
Visual
Minimal 1KB plasma. Brief is barely met.
MiniMax M3 one-shot build of Synthwave — GoldieBench AI benchmark screenshot▶ LIVE
Synthwave
Visual
Neon grid + scanline sun synthwave flythrough.
MiniMax M3 one-shot build of Terrain — GoldieBench AI benchmark screenshot▶ LIVE
Terrain
Visual
Tron-style terrain flythrough.
MiniMax M3 one-shot build of Voxel — GoldieBench AI benchmark screenshot▶ LIVE
Voxel
Visual
29KB Temple-Run-style voxel runner on three.js — lane switching, jump + slide, coins.
MiniMax M3 one-shot build of Waves — GoldieBench AI benchmark screenshot▶ LIVE
Waves
Visual
Gerstner-wave ocean with sun reflection.
every demo, in a grid · click any one to play

Compare MiniMax M3 against every other model

Every head-to-head featuring MiniMax M3. Verdicts shown for scored pairs.

MiniMax M3 vs Fusion
Fusion leads 37–2
MiniMax M3 vs Claude Opus 5
Claude Opus 5 leads 30–17
MiniMax M3 vs Hermes MoA
Hermes MoA leads 31–16
MiniMax M3 vs GPT-5.6 Sol
GPT-5.6 Sol leads 32–15
MiniMax M3 vs Claude Fable 5
Claude Fable 5 leads 26–18
MiniMax M3 vs Qwen 3.8
Qwen 3.8 leads 30–12
MiniMax M3 vs Grok
Grok leads 19–7
MiniMax M3 vs Fugu Ultra
MiniMax M3 leads 17–15
MiniMax M3 vs Kimi K3
Kimi K3 leads 30–16
MiniMax M3 vs GLM-5.2
MiniMax M3 leads 28–11
MiniMax M3 vs Fugu Mini
MiniMax M3 leads 16–15
MiniMax M3 vs Muse Spark 1.2
MiniMax M3 leads 24–22
MiniMax M3 vs Opus 4.8
MiniMax M3 leads 27–14
MiniMax M3 vs Kimi K2.7
MiniMax M3 leads 13–7
MiniMax M3 vs Qwable 5 27B Coder
MiniMax M3 leads 27–13
MiniMax M3 vs Gemini 3.6 Flash
MiniMax M3 leads 29–18
MiniMax M3 vs Claude Sonnet 5
MiniMax M3 leads 28–17
MiniMax M3 vs Qwen 3.7
MiniMax M3 leads 34–4
MiniMax M3 vs Fugu Ultra 1.1
MiniMax M3 leads 17–5
MiniMax M3 vs Inkling
MiniMax M3 leads 44–3
MiniMax M3 vs Agents-A1
MiniMax M3 leads 40–2
MiniMax M3 vs Gemma 4 12B · MLX
MiniMax M3 leads 40–2
MiniMax M3 vs Laguna XS 2.1
MiniMax M3 leads 41–1
MiniMax M3 vs Qwythos 9B
MiniMax M3 leads 42–0
MiniMax M3 vs LongCat-2.0
MiniMax M3 leads 3–0
MiniMax M3 vs Hy3
MiniMax M3 leads 7–0
MiniMax M3 vs Gemma-4 12B Coder
MiniMax M3 leads 6–0
MiniMax M3 vs DeepSeek V4 Flash
47 shared tasks · unscored
MiniMax M3 vs DeepSeek V4 Pro
47 shared tasks · unscored
MiniMax M3 vs Kimi K2.7 · Fast
47 shared tasks · unscored
MiniMax M3 vs Kimi K2.7 · No-Think
47 shared tasks · unscored
MiniMax M3 vs Kimi K2.7 · Quality
47 shared tasks · unscored
MiniMax M3 vs Ornith 1.0
42 shared tasks · unscored
MiniMax M3 vs Claude Mythos 5
Reference-only
MiniMax M3 vs Kilo Code
Reference-only

See all 66 comparisons across every model →

Quick pill index

Direct comparisons against every other scored model on the bench:

MiniMax M3 vs Fusion MiniMax M3 vs Claude Opus 5 MiniMax M3 vs Hermes MoA MiniMax M3 vs GPT-5.6 Sol MiniMax M3 vs Claude Fable 5 MiniMax M3 vs Qwen 3.8 MiniMax M3 vs Grok MiniMax M3 vs Fugu Ultra MiniMax M3 vs Kimi K3 MiniMax M3 vs GLM-5.2 MiniMax M3 vs Fugu Mini MiniMax M3 vs Muse Spark 1.2 MiniMax M3 vs Opus 4.8 MiniMax M3 vs Kimi K2.7 MiniMax M3 vs Qwable 5 27B Coder MiniMax M3 vs Gemini 3.6 Flash MiniMax M3 vs Claude Sonnet 5 MiniMax M3 vs Qwen 3.7 MiniMax M3 vs Fugu Ultra 1.1 MiniMax M3 vs Inkling MiniMax M3 vs Agents-A1 MiniMax M3 vs Gemma 4 12B · MLX MiniMax M3 vs Laguna XS 2.1 MiniMax M3 vs Qwythos 9B MiniMax M3 vs LongCat-2.0 MiniMax M3 vs Hy3 MiniMax M3 vs Gemma-4 12B Coder

Read more on agentos.guide:

MiniMax M3 — frequently asked

What is MiniMax M3?

MiniMax M3 is MiniMax's AI model — 1M-context frontier model at $0.30/M tokens — cheapest big-context model on the bench. It has a 1M tokens context window and was released 2026-06-18.

How good is MiniMax M3 at coding and one-shot builds?

On the GoldieBench one-shot build benchmark it averages 7.97/10 across 47 scored tasks, with 2 gold, 1 silver and 4 bronze medals.

How much does MiniMax M3 cost?

$0.30 / 1M input tokens, $1.50 / 1M output. MiniMax M3 is the cheapest 1M-context frontier model on the bench — roughly 1/200th the per-call cost of OpenRouter Fusion and 1/30th of Claude Opus 4.8. Designed for high-volume agent workloads where context length matt

Where can I see MiniMax M3 demos?

Every one-shot build is live and playable on this page and on the GoldieBench compare matrix — same prompt as every other model, no retries.

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
Join AIPB · $59/mo → Read the Agent OS guides →