Current capability suite
All models
Current published results in a single stream, badged by hardware comparability tier. Tiers come from the exact recorded GPU model. GPU tiers and the hosted subscription reference headline the capability composite; CPU and entry GPU headline timeout-honest usability — open a tier board for rank-comparable numbers. Retired era data remains available only in the source archive.
#
Model
Foundation
Engine
Score
Completed
tok/s
Date
1
Qwen/Qwen3.8-27B-FP8GPU
ndc
vllm
0.96±0.023
0.95 usability
99%
163/165 · 2 timeouts
64.3
Aug 30
2
openai/gpt-oss-120bGPU
ndc
vllm
0.92±0.037
0.91 usability
100%
165/165
132.5
Aug 30
3
google/gemma-4-31B-it-qat-w4a16-ctGPU
Googleqat-w4a16
ndc
vllm
0.92±0.027
0.92 usability
100%
165/165
46.7
Aug 30
4
cyankiwi/Ornith-1.0-35B-AWQ-INT4GPU
cyankiwiawq-int4
ndc
vllm
0.89±0.043
0.89 usability
100%
165/165
94.7
Aug 30
5
poolside/Laguna-S-2.1-NVFP4GPU
ndc
vllm
0.89±0.045
0.89 usability
100%
165/165
98.7
Aug 30
6
deepseek-ai/DeepSeek-V4-Flash-0731GPU
ndc
vllm
0.89±0.053
0.88 usability
100%
165/165
58.5
Aug 30
7
hf.co/unsloth/gpt-oss-20b-GGUF:UD-Q4_K_XLENTRY GPU
OpenAIq4
cdc
ollama
0.76±0.072
0.81 capability
97%
160/165 · 5 timeouts
37.3
Aug 31
8
hf.co/unsloth/Qwen3-Coder-30B-A3B-Instruct-GGUF:Q4_K_MCPU⚠
cdc
ollama
0.67±0.110
0.74 capability
82%
124/152 · 28 timeouts · 10 not attempted (6 h run cap)
14.3
Aug 31
9
hf.co/google/gemma-4-12B-it-qat-q4_0-gguf:latestENTRY GPU⚠
Googleq4
cdc
ollama
0.53±0.129
0.66 capability
60%
99/165 · 66 timeouts · 27 not attempted (6 h run cap)
17.3
Aug 30
10
hf.co/google/gemma-4-26B-A4B-it-qat-q4_0-gguf:latestCPU⚠
Googleq4
cdc
ollama
0.24±0.156
0.27 capability
25%
42/165 · 123 timeouts · 59 not attempted (6 h run cap)
12.6
Aug 30
1
0.96
capability
ndc · vllm99% done · 2 timeouts64.3 tok/s
Categories:
Aug 30
2
openai/gpt-oss-120bGPU
OpenAI0.92
capability
ndc · vllm100% done132.5 tok/s
Categories:
Aug 30
3
0.92
capability
ndc · vllm100% done46.7 tok/s
Categories:
Aug 30
4
0.89
capability
ndc · vllm100% done94.7 tok/s
Categories:
Aug 30
5
poolside/Laguna-S-2.1-NVFP4GPU
poolside0.89
capability
ndc · vllm100% done98.7 tok/s
Categories:
Aug 30
6
deepseek-ai/DeepSeek-V4-Flash-0731GPU
DeepSeek0.89
capability
ndc · vllm100% done58.5 tok/s
Categories:
Aug 30
7
0.76
usability
cdc · ollama97% done · 5 timeouts37.3 tok/s
Categories:
Aug 31
8
0.67
usability
cdc · ollama82% done · 28 timeouts · 10 not attempted (6 h run cap)14.3 tok/s
Categories:
Aug 31
9
0.53
usability
cdc · ollama60% done · 66 timeouts · 27 not attempted (6 h run cap)17.3 tok/s
Categories:
Aug 30
10
0.24
usability
cdc · ollama25% done · 123 timeouts · 59 not attempted (6 h run cap)12.6 tok/s
Categories:
Aug 30
One stream, badged by tier. Each row shows its tier's headline metric, so cross-tier order is context, not a verdict — the tier boards above carry the rank-comparable MDE tiers.