Model comparison
DeepSeek-V2.5 (Sep 2024) vs Grok 4.7
Grok 4.7 is the stronger model overall, scoring 53.1 to 37.6 on the Noometry Index.
Last verified . 14 shared benchmarks.
DeepSeekSummary
- They share 14 benchmarks with published results for both. DeepSeek-V2.5 (Sep 2024) scores higher in 0 categories and Grok 4.7 in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Grok 4.7 leads 62.8 to 34.8.
- DeepSeek-V2.5 (Sep 2024) has downloadable open weights; the other is API-only.
Side by side
| DeepSeek-V2.5 (Sep 2024) | Grok 4.7 | |
|---|---|---|
| Provider | DeepSeek | xAI |
| Noometry Index | 37.6 | 53.1 |
| Released | 2024-09-06 | 2026-09-21 |
| Weights | Open | Proprietary |
| Context window | — | 500K |
| Max output | — | 500K |
| Input $ / M tokens | — | $2 |
| Output $ / M tokens | — | $6 |
| Results tracked | 22 | 39 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok 4.7 leads
DeepSeek-V2.5 (Sep 2024): 31.7 (#281), Grok 4.7: 58.0 (#18)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Grok 4.7 |
|---|---|---|
| LMArena Coding | 1309 | 1427 |
| FrontierCode | — | 47.6% |
| Aider Polyglot | 17.8% | — |
| CursorBench | — | 46.3% |
| LMArena WebDev | — | 1639 |
| FrontierSWE | — | 29.5% |
| SciCode | — | 57.8% |
| BigCodeBench Instruct | 48.6% | — |
| BigCodeBench Complete | 53.2% | — |
| HumanEval+ | 83.5% | — |
| MBPP+ | 74.1% | — |
Agentic & Tool Use Not comparable
DeepSeek-V2.5 (Sep 2024): —, Grok 4.7: 36.7 (#37)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Grok 4.7 |
|---|---|---|
| APEX-Agents | — | 54.6% |
| GDP.pdf | — | 22.8% |
| Vending-Bench 2 | — | 10,537 |
Reasoning Grok 4.7 leads
DeepSeek-V2.5 (Sep 2024): 25.6 (#145), Grok 4.7: 49.1 (#40)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Grok 4.7 |
|---|---|---|
| LMArena Hard Prompts | 1289 | 1413 |
| NYT Connections (extended) | — | 76.8% |
| CritPt | — | 18% |
| Chess Puzzles | — | 38% |
| Mystery Game Puzzles | — | 29% |
| DTBench | — | 96% |
| LMCA | — | 49.4% |
| Epoch Capabilities Index | — | 153.53 |
Math Grok 4.7 leads
DeepSeek-V2.5 (Sep 2024): 35.9 (#177), Grok 4.7: 57.8 (#39)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Grok 4.7 |
|---|---|---|
| LMArena Math | 1288 | 1407 |
| FrontierMath (Tiers 1-3) | — | 53% |
| FrontierMath Tier 4 | — | 17.1% |
| OTIS Mock AIME 2024-2025 | — | 98.1% |
| ProofBench | — | 34% |
Knowledge Grok 4.7 leads
DeepSeek-V2.5 (Sep 2024): 34.8 (#193), Grok 4.7: 62.8 (#22)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Grok 4.7 |
|---|---|---|
| LMArena Expert | 1266 | 1422 |
| GPQA Diamond | — | 92.7% |
| SimpleQA Verified | — | 56% |
Multimodal Not comparable
DeepSeek-V2.5 (Sep 2024): —, Grok 4.7: 35.5 (#87)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Grok 4.7 |
|---|---|---|
| LMArena Vision | — | 1228 |
| Blueprint-Bench 2 | — | 32.5% |
| Furniture Assembly | — | 20.8% |
Multilingual Grok 4.7 leads
DeepSeek-V2.5 (Sep 2024): 42.5 (#193), Grok 4.7: 50.8 (#116)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Grok 4.7 |
|---|---|---|
| LMArena Non-English | 1273 | 1389 |
| LMArena Chinese | 1318 | 1455 |
| LMArena French | 1289 | 1455 |
| LMArena Russian | 1289 | 1397 |
| LMArena Spanish | 1248 | 1400 |
| LMArena German | 1258 | — |
| LMArena Japanese | 1228 | — |
| LMArena Korean | 1209 | — |
Instruction Following Grok 4.7 leads
DeepSeek-V2.5 (Sep 2024): 67.5 (#194), Grok 4.7: 74.1 (#105)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Grok 4.7 |
|---|---|---|
| LMArena Instruction Following | 1280 | 1404 |
Long Context Grok 4.7 leads
DeepSeek-V2.5 (Sep 2024): 39.5 (#174), Grok 4.7: 43.1 (#104)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Grok 4.7 |
|---|---|---|
| LMArena Longer Query | 1301 | 1413 |
Writing & Preference Grok 4.7 leads
DeepSeek-V2.5 (Sep 2024): 49.8 (#187), Grok 4.7: 70.0 (#24)
| Benchmark | DeepSeek-V2.5 (Sep 2024) | Grok 4.7 |
|---|---|---|
| LMArena Text | 1294 | 1399 |
| LMArena Creative Writing | 1285 | 1391 |
| LMArena Multi-Turn | 1297 | 1393 |
| EQ-Bench Creative Writing | — | 2007 |
Frequently asked questions
Is DeepSeek-V2.5 (Sep 2024) better than Grok 4.7?
Grok 4.7 is the stronger model overall, scoring 53.1 to 37.6 on the Noometry Index.
Is DeepSeek-V2.5 (Sep 2024) or Grok 4.7 better for coding?
Grok 4.7 scores higher on coding benchmarks: 58.0 versus 31.7 in the Noometry coding category.
How many benchmarks do DeepSeek-V2.5 (Sep 2024) and Grok 4.7 share?
14 benchmarks have published results for both models. DeepSeek-V2.5 (Sep 2024) has 22 scored results on Noometry and Grok 4.7 has 39.