Tencent / model / guest in this test
Hy4 (full)
AngelSlim IQ2 GGUF, about 3.5 bits per weight, revision c65d2c9, running on Mac Studio M3 Ultra, 512 GB.
Guest, not a fair third contestant: different machine, different runtime, heavier compression.
Highlights
Hy4 full in Qwen3.8 Flash Next vs Hy4 Flash
From its roasts
Attempt 1, in fullAnd 3.8 — not 3, not 4, 3.8. That's not a version, that's a Yelp rating for a dim sum place doing fine.
A game it wrote

These games were written for a keyboard and a big screen. On a phone they may not respond. They are best on a computer.
Test 01 / Pilot
Results in Qwen3.8 Flash Next vs Hy4 Flash
The full Hy4 is not a fair third contestant: different machine, different runtime, much heavier compression. Read its column as what the big model does on a Mac, not as a ranking against the other two.
Pelican on a bicycle
Opinion
- Valid drawings, of 3 attempts
- 3
Octopus and penguin band
Opinion
- Valid drawings, of 3 attempts
- 3
Playable browser game
Automatic load test; the rest is opinion
- Complete games
- 2 of 3 (1 cut off at the answer limit)
Roast
Opinion
3.8 — not 3, not 4, 3.8. That's not a version, that's a Yelp rating for a dim sum place doing fine.
You didn't open-source a model, you opened a warehouse club.
- My ranking (opinion)
- Place 1 of 3
Text adventure
Rules check plus opinion
- Complete games
- 0 of 3: empty answer on turn 2, probably a runtime fault, unconfirmed
Attempt 1 (empty answer)Attempt 2 (empty answer)Attempt 3 (empty answer)
Bug fix
Hidden tests, automatic
- Bugs fixed of 3, per attempt
- 3, 3, 3
- Minutes per attempt
- over 60
Contract questions
Answer key
- Score of 5, per attempt
- 5, 5, 5
- Slip
- None
12 steps with a contradiction
Answer key
- Flagged the contradiction
- 2 of 3
- Arithmetic right
- 3 of 3
Network design
Opinion
- Words per plan
- 3,200 to 3,800
- Access points proposed
- 6 to 11
Every attempt, with time and tokens
| Challenge | Attempt | Status | Minutes | Tokens |
|---|---|---|---|---|
| Playable browser game | 1 | Finished | 86.3 | 57,749 |
| Playable browser game | 2 | Cut off at the answer limit | 90.1 | 60,000 |
| Playable browser game | 3 | Finished | 35.5 | 24,737 |
| Bug fix | 1 | Finished | 70.8 | 47,830 |
| Bug fix | 2 | Finished | 78.4 | 52,692 |
| Bug fix | 3 | Finished | 68.8 | 46,635 |
| Network design | 1 | Finished | 66.4 | 45,093 |
| Network design | 2 | Finished | 55.4 | 37,529 |
| Network design | 3 | Finished | 65.8 | 44,274 |
| Contract questions | 1 | Finished | 64.2 | 42,673 |
| Contract questions | 2 | Finished | 31.4 | 21,004 |
| Contract questions | 3 | Finished | 36.6 | 24,577 |
| 12 steps with a contradiction | 1 | Finished | 20.6 | 14,331 |
| 12 steps with a contradiction | 2 | Finished | 22.5 | 15,718 |
| 12 steps with a contradiction | 3 | Finished | 19.9 | 13,948 |
| Pelican on a bicycle | 1 | Finished | 29.0 | 20,364 |
| Pelican on a bicycle | 2 | Finished | 27.0 | 18,860 |
| Pelican on a bicycle | 3 | Finished | 22.9 | 16,007 |
| Octopus and penguin band | 1 | Finished | 31.1 | 21,474 |
| Octopus and penguin band | 2 | Finished | 34.5 | 23,884 |
| Octopus and penguin band | 3 | Finished | 27.7 | 19,745 |
| Roast | 1 | Finished | 7.1 | 5,177 |
| Roast | 2 | Finished | 5.9 | 4,313 |
| Roast | 3 | Finished | 8.4 | 6,079 |
| Text adventure | 1 | Empty answer | 12.2 | 8,665 |
| Text adventure | 2 | Empty answer | 15.0 | 10,517 |
| Text adventure | 3 | Empty answer | 15.6 | 10,949 |
Build and settings
What exactly ran
- Maker
- Tencent
- Build
- AngelSlim IQ2 GGUF, about 3.5 bits per weight, revision c65d2c9
- File size
- 310 GB
- Machine
- Mac Studio M3 Ultra, 512 GB
- Runtime
- llama.cpp PR #32517 at f8d4cc4, Metal
- Sampling
- temperature 0.9, top-p 1.0 (maker settings)
- Thinking
- on (default)
Speed
How fast it ran here
- Writing speed, short prompt
- 12.2 tok/s
- Writing speed after a 32K prompt
- 10.9 tok/s
- Writing speed after a 127K prompt
- 10.0 tok/s
- Reading speed, 32K prompt
- 79 tok/s
- Reading speed, 127K prompt
- 75 tok/s
- Wait for first token, 32K prompt
- 404 s
- Wait for first token, 127K prompt
- 1,680 s (28 min)
- Load time
- 13 s (warm)
- Load time, cold
- 123 s (2.0 min)
- Least memory left free
- 139 GiB
- Time for all 27 attempts
- 17.5 hours
- Tokens generated, all attempts
- 714,824 tokens
- Short-prompt writing speed is the average of three runs; long-prompt rows are one run each.
- Hy4 Flash speed rows come from the first pilot night (same files and runtime); Qwen rows from the rerun on the current build.
- Speed tests use one fixed sampling setting for every model.
- About nine tenths of what all three models wrote was thinking, not answer.