Best model for famous gotchas
Models ranked by their famous gotchas score, 0 to 100, as on the category page. Only ranked models with enough scored items are included, and models are grouped by total parameters.
Read the memory figure as an estimate, not a guarantee. The memory a model needs is estimated from its file: the weights plus about 20% for the context of a single conversation. It is not a measurement, and a long conversation or several at once needs more. A model with no file size on record is not counted as fitting any amount of memory. Speeds, and the GPU memory shown on the speed page, were measured on this benchmark's server, not on yours. That server kept 1 to 8 answers in flight at a time and loaded a context window of 16,384 tokens for most models; 42 of the models ranked ran at a shorter context. Speeds were measured on one laptop and are comparable only between models on the same backend. See the settings and the memory method.
All sizes
gemma-4-26B-A4B-it leads here with 100.0 on a scale of 0 to 100; its Overall is 97.8 (95% interval 95.8–99.3).
| Place | Model | Famous gotchas (0 to 100) | Overall | Rank range | Decode (tok/s) | Memory needed (GB, est.) | Size class |
|---|---|---|---|---|---|---|---|
| 1 | gemma-4-26B-A4B-it | 100.015 of 15 items scored | 97.895.8–99.3 | 1–22 | 15.7ROCm 7.1 | 21.6 | 16–35B |
| 1 | gpt-oss-120b | 100.015 of 15 items scored | 98.696.8–100.0 | 1–14 | 11.0Vulkan | 78.4 | > 35B |
| 1 | gpt-oss-20b | 100.015 of 15 items scored | 95.092.2–97.6 | 2–60 | 14.9ROCm 7.1 | 16.6 | 16–35B |
| 1 | Muse Glimmer 30B | 100.015 of 15 items scored | 97.996.0–99.3 | 1–23 | 3.6ROCm 7.1 | 21.8 | 16–35B |
| 1 | Nemotron 3.5 Lightning 30B A3B | 100.015 of 15 items scored | 92.889.7–95.8 | 5–64 | 17.7ROCm 7.1 | 30.5 | 16–35B |
Tiny (under 3B)
Exaone Deep 2.4B leads its size class with 93.3 on a scale of 0 to 100; its Overall is 54.7 (95% interval 48.9–60.5).
| Place | Model | Famous gotchas (0 to 100) | Overall | Rank range | Decode (tok/s) | Memory needed (GB, est.) | Size class |
|---|---|---|---|---|---|---|---|
| 1 | Exaone Deep 2.4B | 93.315 of 15 items scored | 54.748.9–60.5 | 156–196 | 31.2ROCm 7.1 | 1.9 | Tiny (under 3B) |
| 1 | Lfm3 1.7B | 93.315 of 15 items scored | 70.265.3–75.3 | 96–162 | 42.1ROCm 7.1 | 1.3 | Tiny (under 3B) |
| 3 | Deepcoder 1.5B Previewfine-tune of DeepSeek-R1-Distill-Qwen 1.5Bcode model | 80.015 of 15 items scored | 55.850.3–61.0 | 155–192 | 44.9ROCm 7.1 | 1.3 | Tiny (under 3B) |
| 3 | Lfm2.5 Thinking 1.2B | 80.015 of 15 items scored | 74.369.5–79.1 | 84–148 | 205.0ROCm 7.1 | 0.9 | Tiny (under 3B) |
| 3 | Lfm3 VL 2B | 80.015 of 15 items scored | 72.267.2–77.0 | 89–158 | 41.9ROCm 7.1 | 1.8 | Tiny (under 3B) |
3–4B
Phi4 Mini Reasoning 3.8B leads its size class with 93.3 on a scale of 0 to 100; its Overall is 70.9 (95% interval 65.9–76.0).
| Place | Model | Famous gotchas (0 to 100) | Overall | Rank range | Decode (tok/s) | Memory needed (GB, est.) | Size class |
|---|---|---|---|---|---|---|---|
| 1 | Phi4 Mini Reasoning 3.8B | 93.315 of 15 items scored | 70.965.9–76.0 | 97–157 | 48.8ROCm 7.1 | 3.8 | 3–4B |
| 2 | Granite4.2 3B | 86.715 of 15 items scored | 78.273.7–83.1 | 72–140 | 24.3ROCm 7.1 | 2.7 | 3–4B |
| 3 | Granite4 3B | 80.015 of 15 items scored | 74.169.0–79.2 | 85–145 | 71.3ROCm 7.1 | 2.5 | 3–4B |
| 3 | Phi-3.5-mini-instruct | 80.015 of 15 items scored | 67.162.0–72.3 | 110–174 | 17.6ROCm 7.1 | 2.9 | 3–4B |
| 3 | Phi-4-mini-instruct | 80.015 of 15 items scored | 81.677.1–86.0 | 62–123 | 17.1ROCm 7.1 | 3.0 | 3–4B |
4–9B
Qwen3-4B leads its size class with 100.0 on a scale of 0 to 100; its Overall is 89.2 (95% interval 85.6–92.4).
| Place | Model | Famous gotchas (0 to 100) | Overall | Rank range | Decode (tok/s) | Memory needed (GB, est.) | Size class |
|---|---|---|---|---|---|---|---|
| 1 | Qwen3-4B | 100.015 of 15 items scored | 89.285.6–92.4 | 18–81 | 17.0ROCm 7.1 | 3.1 | 4–9B |
| 2 | Exaone Deep 7.8B | 93.315 of 15 items scored | 65.760.0–71.5 | 120–170 | 12.2ROCm 7.1 | 5.8 | 4–9B |
| 2 | Gemma4 E2B IT | 93.315 of 15 items scored | 93.891.0–96.1 | 4–61 | 17.2ROCm 7.1 | 5.5 | 4–9B |
| 2 | Gemma4 E4B IT | 93.315 of 15 items scored | 95.593.0–97.7 | 2–51 | 13.2ROCm 7.1 | 7.9 | 4–9B |
| 2 | Nemotron 3.5 Bolt E4B | 93.315 of 15 items scored | 90.386.3–93.9 | 8–82 | 9.8ROCm 7.1 | 8.6 | 4–9B |
9–16B
Ornith-1.5-9B leads its size class with 100.0 on a scale of 0 to 100; its Overall is 89.0 (95% interval 85.4–92.4).
| Place | Model | Famous gotchas (0 to 100) | Overall | Rank range | Decode (tok/s) | Memory needed (GB, est.) | Size class |
|---|---|---|---|---|---|---|---|
| 1 | Ornith-1.5-9B | 100.015 of 15 items scored | 89.085.4–92.4 | 19–81 | 8.1ROCm 7.1 | — | 9–16B |
| 2 | DeepSeek-R1-Distill-Qwen-14B | 93.315 of 15 items scored | 85.381.2–89.2 | 38–106 | 7.1ROCm 7.1 | 10.8 | 9–16B |
| 2 | Gemma 4 12B IT | 93.315 of 15 items scored | 98.697.0–99.8 | 1–12 | 8.1ROCm 7.1 | 8.9 | 9–16B |
| 2 | Qwen3.5-9B | 93.315 of 15 items scored | 92.089.0–95.1 | 11–69 | 8.4ROCm 7.1 | 6.8 | 9–16B |
| 5 | Falcon3 10B Instruct | 86.715 of 15 items scored | 81.376.6–85.8 | 57–133 | 9.0ROCm 7.1 | 7.6 | 9–16B |
16–35B
gemma-4-26B-A4B-it leads its size class with 100.0 on a scale of 0 to 100; its Overall is 97.8 (95% interval 95.8–99.3).
| Place | Model | Famous gotchas (0 to 100) | Overall | Rank range | Decode (tok/s) | Memory needed (GB, est.) | Size class |
|---|---|---|---|---|---|---|---|
| 1 | gemma-4-26B-A4B-it | 100.015 of 15 items scored | 97.895.8–99.3 | 1–22 | 15.7ROCm 7.1 | 21.6 | 16–35B |
| 1 | gpt-oss-20b | 100.015 of 15 items scored | 95.092.2–97.6 | 2–60 | 14.9ROCm 7.1 | 16.6 | 16–35B |
| 1 | Muse Glimmer 30B | 100.015 of 15 items scored | 97.996.0–99.3 | 1–23 | 3.6ROCm 7.1 | 21.8 | 16–35B |
| 1 | Nemotron 3.5 Lightning 30B A3B | 100.015 of 15 items scored | 92.889.7–95.8 | 5–64 | 17.7ROCm 7.1 | 30.5 | 16–35B |
| 1 | Olmo 3.1 32B Instruct | 100.015 of 15 items scored | 89.686.0–93.2 | 10–82 | 3.1ROCm 7.1 | 23.4 | 16–35B |
> 35B
gpt-oss-120b leads its size class with 100.0 on a scale of 0 to 100; its Overall is 98.6 (95% interval 96.8–100.0).
| Place | Model | Famous gotchas (0 to 100) | Overall | Rank range | Decode (tok/s) | Memory needed (GB, est.) | Size class |
|---|---|---|---|---|---|---|---|
| 1 | gpt-oss-120b | 100.015 of 15 items scored | 98.696.8–100.0 | 1–14 | 11.0Vulkan | 78.4 | > 35B |
| 1 | Qwen3-Coder-Nextcode model | 100.015 of 15 items scored | 96.494.2–98.3 | 2–42 | 18.1Vulkan | 62.1 | > 35B |
| 3 | GLM-4.5-Air | 93.315 of 15 items scored | 92.889.7–95.6 | 6–68 | 6.0Vulkan | 87.6 | > 35B |
| 3 | Llama 4 Scout 17B-16E Instruct | 93.315 of 15 items scored | 93.690.7–96.1 | 4–65 | 17.0Vulkan | 80.9 | > 35B |
| 3 | Llama-3.3-70B-Instruct | 93.315 of 15 items scored | 94.391.3–96.7 | 4–63 | 1.4Vulkan | 51.0 | > 35B |
A higher place is not always a clear win. When two models have overlapping rank ranges, this battery cannot tell them apart; see how sure a score is. A dash means not measured or not scored, which is different from 0. Speeds were measured on one laptop and are comparable only between models on the same backend.
More: all the best-model pages, the hardware picker, speed and memory for every model.