qwen3-4b-instruct-2507
| Released | 2025-08-06 |
|---|---|
| Price per 1M | Not listed |
| Overall score | 38.7 (early, few results) |
| Provider | Unknown |
| Weights | Unknown |
Scores
| Dimension | Score | Rank | Benchmarks |
|---|---|---|---|
| Overall: Balanced | 38.7 | 3 | |
| Overall: Coding | 36.9 | 2 | |
| Overall: Reasoning | 40.1 | 2 | |
| Overall: Agentic work | 37.0 | 2 | |
| Agents | 37.3 | 1 | |
| Reasoning | 35.9 | 1 | |
| Math | 45.5 | 1 |
Evidence
Agents 37.3 ±3.6
Reasoning 35.9 ±4.6
| Benchmark | Result | Source | Date |
|---|---|---|---|
| GPQA Diamond | 45.8% | Epoch AI | 2026-08-28 |
Math 45.5 ±3.6
| Benchmark | Result | Source | Date |
|---|---|---|---|
| OTIS Mock AIME 2024-2025 | 52.2% | Epoch AI | 2026-08-28 |
Results not used in the score
| Benchmark | Result | Source | Why not used |
|---|---|---|---|
| BFCL v4 | 35.5% | BFCL | A stronger configuration result was used |
Names matched to this model
qwen3-4b-instruct-2507