Run Detail
completedglm-5.3-flashRecorded for these videos
27:18Gemini 4 Hydrogen: Cheaper Tokens, More Expensive Answers?
Nov 13, 2026 · YouTube
18:40ERNIE 5.5 Flash Preview vs GLM 5.3 Flash: Baidu's new model is very impressive!
Nov 11, 2026 · YouTube
36:22Fable 5.2 robbing me blind, plus: Nex AGI's N2.6 vs GLM 5.3 Flash
Nov 11, 2026 · YouTube
14:03Gemini 4 Carbon: Can it handle my challenges? (First look with one-shot coding tests)
Nov 10, 2026 · YouTube
47:51Ranking major U.S. Open-Weights Models. (Where does Ripple 65B fit in?)
Nov 6, 2026 · YouTube
10:57"Crouton Alpha": the latest super-efficient coding model by Meta?
Nov 3, 2026 · YouTube
38:58Testing Particle-1: a post-trained DeepSeek V4.1 Flash specializing in coding and efficiency
Oct 28, 2026 · YouTube
22:14GPT-6.1 Terra Pro vs Terra: How Much Better Is Pro?
Oct 27, 2026 · YouTube
29:55Kimi K3.1 Pro Looked Incredible. Then I Kept Testing.
Oct 26, 2026 · YouTube
31:28Is this company reselling GPT Sol 6.1 at a loss (for non-agentic/text-only)?🤔 Is this sustainable?
Oct 20, 2026 · YouTube
17:58Nemotron 3.5 Super VL vs Opus 5.5: Testing NVIDIA’s new coding juggernaut
Oct 15, 2026 · YouTube
16:04Testing a *massive* new open-weights model 🤯: Muse Spark Open (vs. Nemotron, Opus 5.5, and more)
Oct 15, 2026 · YouTube
16:43Gemini 4 Argon: Great 🐈 Animation, but a broken escalator. (Coding challenge vs Opus 5.5, Sol 6.1)
Oct 13, 2026 · YouTube
16:37Meta's new "Muse Wonder" AI is impressive... but can it compete with Anthropic's models?
Oct 12, 2026 · YouTube
11:12WOW... Haiku 5.5 is very impressive. RIP OpenAI?
Oct 7, 2026 · YouTube
22:47Mistral Large 4 Is Cheap. But Is It Good Enough to Compete?
Oct 6, 2026 · YouTube
22:02Is Pareto 26.10 Preview Just Routing to GPT Sol? I Tested It Against Sol 6.1 Pro
Oct 1, 2026 · YouTube
11:22I Tested Sonnet 5.5 Against Opus 5.5. Is It Actually Cheaper?
Sep 28, 2026 · YouTube
30:00Did Elon Musk's team "train the gay away"... and make Grok 4.7 worse?
Sep 26, 2026 · YouTube
19:21I Tested the Stealth "Space Bunny Alpha" Model Before It Went Offline
Sep 23, 2026 · YouTube
10:58So... how much better is Claude Opus 5.5 vs. GPT-6 Sol, really?
Sep 23, 2026 · YouTube
9:38Claude Opus 5.5... the BEST new model for coding!! 🤩
Sep 22, 2026 · YouTube
29:00Grok 4.7: improved, but GLM 5.3 FlashX beat it anyway (benchmark against Gemini, Sol, etc.)
Sep 22, 2026 · YouTube
16:54Stealth Union Alpha: Is this another cheap GPT model? (better than Luna, cheaper than Terra?)
Sep 16, 2026 · YouTube
16:45DeepSeek V4.1 Flash Is Shockingly Good for the Price
Sep 10, 2026 · YouTube
17:14How much better is GPT-6 Astra over Claude Opus 5? Real Website and Coding Tests
Sep 5, 2026 · YouTube
29:13Implementing a multiplayer game with GPT-6 Astra (+ Gemini 3.8, Opus 5, Sol Pro)
Sep 5, 2026 · YouTube
19:42Meta Muse Spark 1.3 Review: A Huge Leap Forward
Sep 2, 2026 · YouTube
17:31Testing Gemini 3.8 Flash on Real Coding Challenges (vs. other models)
Sep 2, 2026 · YouTube
14:09Testing IBM Granite-4.2-8B! Can it name 100 women?
Aug 31, 2026 · YouTube
17:37Tencent Hy4 Preview Tested: Is It Really Better Than Kimi K3 and GLM 5.3?
Aug 28, 2026 · YouTube
14:50Stealth Ox Alpha: a new GLM model! Full test (coding & writing) against Grok, Gemini, GPT & others
Aug 21, 2026 · YouTube
21:54GLM 5.3 Tested: Does it beat Claude Opus 5 & Grok 4.6?
Aug 19, 2026 · YouTube
21:23Giving Qwen 3.8 27B another chance (+ GLM 5.2). Testing against GPT, Gemini, Grok etc.
Aug 18, 2026 · YouTube
51:00AI Coding Showdown: Grok 4.6, Gemini 3.7, Terra-Pro, Qwen3.8 2.4TB, Muse Glimmer & more
Aug 15, 2026 · YouTube
Provider
openrouter
Challenge Version
v2.0
Turns
1
Duration
585.3s
Total Tokens
104,483
Input Tokens
638
Output Tokens
103,845
Reasoning Tokens
102,574
Reasoning Effort
default
Est. Cost
$0.0520
Started
11/10/2026, 3:37:59 PM
OR Provider
Together
Finish Reason
stop
Generation ID
gen-1794325078-wc829QUMKcQfj2UXnVMU
Turns (1)
Turn #1
638 in103,845 out102,574 reasoning104,483 total585.3s
user
You are a size coder. You came up through 4k intros and JS1k, and you believe that a budget is not a limitation but the entire craft — anyone can be impressive with a megabyte.
I want something worth watching, and I am giving you 2048 bytes to do it in.
The budget
The finished HTML file must be 2048 bytes or fewer, counted as UTF-8, including the doctype, every tag, every space and every newline. That is the whole file. There is no second file.
House rules
- One file. Nothing external — no CDN, no stylesheet, no font, no script, no image.
data:URIs are allowed but they count against you like everything else. - No network at runtime. This is opened offline.
- No
localStorage,sessionStorage, cookies orindexedDB. It runs in a sandboxed frame where touching them throws. - No build step. Saved to disk and double-clicked, it runs.
- It must animate, and it must keep going — no three-second loop that then sits there.
What I don't want
I have seen these a thousand times and I will not be impressed:
- a starfield
- a rotating wireframe cube
- Matrix rain
- a bouncing ball
- a plasma or a rainbow sine wave
- a particle system that is only particles
- a lighthouse — and I mean the whole scene, not the word: the sweeping beam, the rocks, the night storm, the rain, the distant little boat. Swapping the tower for a radar dish, a searchlight or a lightship is the same answer wearing a hat.
Show me something with a subject. Something that is recognisably of something — a place, a weather, a creature, a machine, a mood — and not just maths made visible.
Answer in exactly these sections
The pitch
One line. What am I about to see when I open this file?
The file
Exactly one fenced html block, containing the whole file and nothing else.
The ledger
- Bytes: the exact byte count of the file above.
- How you counted: in one line.
- Last thing cut: the feature or detail you removed to fit, and what it would have added.
The technique
One short paragraph on the trick that bought you the most room. Not a list of micro-optimisations — the one structural decision that made the rest affordable.
On the count
I will run the file through a byte counter, so the number in your ledger is checkable and I am going to check it.
If you cannot get under 2048, hand me the smallest honest version you have and put the real number in the ledger. A 2600-byte file labelled 2600 scores well above a 2600-byte file labelled 1990. Going over is a miss. Lying about going over is the only thing here that is disqualifying.
assistant
The pitch
Moon jellyfish pulsing up through black water — translucent bells with their four pink gonad rings showing through, tentacles trailing, light shafts from the surface, marine snow sinking past.
The file
<!doctype html><title>Drift</title><canvas id=c><script>
C=c.getContext`2d`,M=Math,P=M.PI,z=0,R=requestAnimationFrame,B=_=>C.beginPath(),T=(a,b)=>C.lineTo(a,b),S=_=>C.stroke()
J=[[.2,.4,.6,.8,2,.5],[.85,.9,.45,1.2,4,.4],[.6,.35,.4,.9,6,.3],[.5,.7,1.4,1,0,1]]
function D(x,y,s,ph,a){
p=M.sin(ph),w=60*s-9*s*p,h=46*s+9*s*p
C.strokeStyle='rgba(255,190,235,'+.4*a+')',C.lineWidth=1
for(i=0;i<14;i++){b=x-w+i/6.5*w,B(),C.moveTo(b,y),v=h*(1.8+M.sin(i*2.4))
for(k=1;k<7;k++)T(b+(M.sin(t*2.2+i*.7+k)*4+p*2)*k*s,y+v*k/6)
S()}
C.strokeStyle='rgba(255,150,220,'+.35*a+')',C.lineWidth=4*s
for(i=0;i<4;i++){b=x+(i-1.5)*w*.25,B(),C.moveTo(b,y)
for(k=1;k<6;k++)T(b+M.sin(t*1.8+i*2+k*1.4)*k*6*s,y+1.6*h*k/5)
S()}
C.shadowColor='rgba(255,140,220,'+.45*a+')',C.shadowBlur=45*s
B(),C.ellipse(x,y,w,h,0,P,0)
for(k=0;k<13;k++)T(x+w-w*k/6,y+M.abs(M.sin(k*1.57))*5*s)
C.fillStyle='rgba(225,150,255,'+.22*a+')',C.fill()
C.strokeStyle='rgba(255,205,250,'+.5*a+')',C.lineWidth=1.5,S()
C.shadowBlur=0
C.strokeStyle='rgba(255,110,190,'+.55*a+')',C.lineWidth=2.5*s
for(i=0;i<4;i++)B(),C.ellipse(x+(i%2*2-1)*.23*w,y-h*(.43-.11*(i>1)),w*.15,h*.11,0,0,7),S()
C.strokeStyle='rgba(255,190,240,'+.25*a+')',C.lineWidth=1,B()
for(i=0;i<8;i++)A=P*(1+i/7),C.moveTo(x,y-h*.9),T(x+M.cos(A)*w*.75,y+M.sin(A)*h*.7)
S()}
function L(n){t=n/1e3
W=c.width=innerWidth,H=c.height=innerHeight
g=C.createLinearGradient(0,0,0,H),g.addColorStop(0,'#0b2e40'),g.addColorStop(1,'#020409'),C.fillStyle=g,C.fillRect(0,0,W,H)
C.fillStyle='rgba(150,220,255,.05)',B()
for(i=0;i<4;i++)x=i*W/4+M.sin(t*.14+i)*80,C.moveTo(x,0),T(x+70,H),T(x+150,H)
C.fill()
C.fillStyle='rgba(210,235,255,.3)'
for(i=0;i<80;i++)C.fillRect((i*2657+t*7)%W,(i*7577+t*30*(1+i%3))%H,2,2)
for(j of J){ph=t*2*j[3]+j[4],x=j[0]*W+M.sin(t*.35+j[4])*40*j[2],j[1]-=j[2]*(.4+1.2*M.max(0,M.sin(ph))**2)*(n-z)/16.7,j[1]<-260&&(j[1]=H+50),D(x,j[1],j[2],ph,j[5])}z=n
R(L)}
R(L)
</script>
The ledger
- Bytes: 1886
- How you counted: Character-by-character tally of the file exactly as written — pure ASCII, so UTF-8 bytes equal characters — 1851 characters across 36 lines plus 35 newlines, no trailing newline, no BOM.
- Last thing cut: The bell's gradient fill and a second inner-dome highlight pass, replaced by a single
shadowBlurhalo on the outline. It would have given the bell a lit inner surface for ~120 bytes I preferred to hold as honest margin.
The technique
The structural decision that paid for everything else: one parameterised animal. D(x, y, scale, phase, alpha) draws a complete jellyfish — glow, tentacle fringe, oral arms, scalloped bell, the four-ring gonad clover, radial canals — at any size and depth, so four creatures on screen cost a single code path and the small dim ones are a few bytes of data in an array. Everything that moves keys off that one phase value: the bell narrows and stretches on contraction, thrust is gated on it, tentacle drag bends with it, so the swimming reads as behaviour rather than wiggle, with no extra state. Even the marine snow is stateless — each flake is just the loop index hashed through modular multiplication against the viewport — so there is no particle array to store or step, and the whole scene rebuilds from six numbers per jelly per frame.
Attachments
z-ai/glm-5.3-flash — page