Qwen3-Next-80B-A3B-Thinking vs Edge0-120B-NVFP4

New Comparison

Scores are fetched live from the current database on every visit — nothing is cached.

Benchmark Scores Comparison

Qwen3-Next-80B-A3B-Thinking Edge0-120B-NVFP4

Qwen3-Next-80B-A3B-Thinking

80.0B params

Score: 100.0

Edge0-120B-NVFP4

120.0B params

Coverage matrix

2 models × 29 benchmarks — hover a benchmark name for its category

Full Coverage Matrix
Hides benchmarks where all selected models score identically.
Shading = score relative to the best score among the selected models: Low Mid High Missing (—) The totals row shows how many of the 29 benchmarks each model has scores for.
Benchmark Qwen3-Next-80B-A3B-Thinking Qwen Edge0-120B-NVFP4 Edge0
HMMT Feb 25 73.9 —
GPQA Diamond — 87.4
TauBench V2 Telecom 43.9 —
LiveCodeBench v6 68.7 —
AA-LCR — 81
Tau-Bench Retail 69.6 —
Multi-IF 77.8 —
OJBench 29.7 —
TauBench V2 Retail 67.8 —
INCLUDE 78.9 —
PolyMATH 56.3 —
MMLU-ProX 78.7 —
MMLU-Pro 82.7 —
SuperGPQA 60.8 —
IFEval 88.9 —
HLE (with tools) — 31
MMLU-Redux 92.5 —
IFBench — 79.3
WritingBench 84.6 —
LiveBench 241125 76.6 —
Terminal-Bench 2.1 (Terminus-2) — 75.3
Arena-Hard v2 62.3 —
GPQA 77.2 —
AIME 2025 87.8 —
Tau-Bench Airline 49 —
TauBench V2 Airline 60.5 —
DeepSWE 1.1 — 51.3
BFCL-V3 72 —
CFEval 2071 —
Covered 23/29 6/29

Scorecard

Overall 0–100 score — category strengths (50%), shared-benchmark wins (30%), coverage (20%)

Model Overall score Shared benchmark wins Coverage Strongest at Weakest at
Qwen3-Next-80B-A3B-Thinking 79.3 — 23 of 29 (79.0%) — —
Edge0-120B-NVFP4 20.7 — 6 of 29 (21.0%) — —

Partial coverage benchmarks

Scored by at least one but not all selected models (29) — missing scores show as —

Benchmark Category Qwen3-Next-80B-A3B-Thinking Edge0-120B-NVFP4
HMMT Feb 25 1/2 models stem_reasoning 73.9 —
GPQA Diamond 1/2 models stem_reasoning — 87.4
TauBench V2 Telecom 1/2 models general_agent 43.9 —
LiveCodeBench v6 1/2 models stem_reasoning 68.7 —
AA-LCR 1/2 models long_context — 81
Tau-Bench Retail 1/2 models general_agent 69.6 —
Multi-IF 1/2 models instruction_following 77.8 —
OJBench 1/2 models stem_reasoning 29.7 —
TauBench V2 Retail 1/2 models general_agent 67.8 —
INCLUDE 1/2 models multilingual 78.9 —
PolyMATH 1/2 models multilingual 56.3 —
MMLU-ProX 1/2 models multilingual 78.7 —
MMLU-Pro 1/2 models knowledge 82.7 —
SuperGPQA 1/2 models knowledge 60.8 —
IFEval 1/2 models instruction_following 88.9 —
HLE (with tools) 1/2 models stem_reasoning — 31
MMLU-Redux 1/2 models knowledge 92.5 —
IFBench 1/2 models instruction_following — 79.3
WritingBench 1/2 models instruction_following 84.6 —
LiveBench 241125 1/2 models reasoning 76.6 —
Terminal-Bench 2.1 (Terminus-2) 1/2 models coding_agent — 75.3
Arena-Hard v2 1/2 models instruction_following 62.3 —
GPQA 1/2 models reasoning 77.2 —
AIME 2025 1/2 models stem_reasoning 87.8 —
Tau-Bench Airline 1/2 models general_agent 49 —
TauBench V2 Airline 1/2 models general_agent 60.5 —
DeepSWE 1.1 1/2 models coding_agent — 51.3
BFCL-V3 1/2 models general_agent 72 —
CFEval 1/2 models coding_agent 2071 —

What now?

You've seen 2 models side by side — keep the loop going.

Share this comparison

The URL is the comparison — copy or share it anywhere. No tracking, no cookies.

Not the right matchup?

Swap one model out — the pre-filled form keeps the rest of the selection.

Swap Qwen3-Next-80B-A3B-Thinking × Swap Edge0-120B-NVFP4 ×