DeepSeek V4.1 Flash vs Grok 4.7
| Spec | DeepSeek V4.1 Flash | Grok 4.7 |
|---|---|---|
| Respan Index | 52.5 (#37) | 53.0 (#34) |
| Blended price per 1M | $0.089 | $3 |
| Context window | 1M | 500K |
| Input / output per 1M | $0.051 / $0.203 | $2 / $6 |
| Max output | 384K | - |
| Input | text, vision | text, vision |
| Released | 2026-09-10 | 2026-09-21 |
Respan Index by category
Each category is scored 0 to 100, and how far ahead the winner is. Leads under a point are close to even.
| Category | DeepSeek V4.1 Flash | Grok 4.7 |
|---|---|---|
| Coding | 43.7 DeepSeek V4.1 Flash | 42.6 Grok 4.7 |
| Agents | 58.4 DeepSeek V4.1 Flash | 70.4 Grok 4.7 |
| Reasoning | 55.5 DeepSeek V4.1 Flash | 42.7 Grok 4.7 |
| Knowledge & language | 44.5 DeepSeek V4.1 Flash | 55.5 Grok 4.7 |
| Human preference | 57.5 DeepSeek V4.1 Flash | 56.9 Grok 4.7 |
Benchmarks both report
DeepSeek V4.1 Flash leads on 17 of the 25 benchmarks both report. Higher is better. Each score links to where it was published.
| Benchmark | DeepSeek V4.1 Flash | Grok 4.7 |
|---|---|---|
| LMArena WebDev (LMArena) | 1658 | 1639 |
| LMArena Elo | 1475 | 1443 |
| LMArena Vision (LMArena) | 1262 | 1250 |
| LMArena Agent (LMArena) | 0.0078 | 0.0148 |
| MirrorCode (Epoch AI) | 64.4% | 54.4% |
| APEX-Agents (Mercor) | 39.5% | 70% |
| FrontierMath Tier 4 v2 (Epoch AI) | 26.8% | 17.1% |
| FrontierMath Tiers 1-3 v2 (Epoch AI) | 67.4% | 53% |
| GPQA Diamond | 90.9% | 92.7% |
| Mystery Game Puzzles (Epoch AI) | 43% | 29% |
| OTIS Mock AIME 2024-2025 (Epoch AI) | 98.3% | 98.1% |
| Terminal-Bench 4.0 (Vals AI) | 19.7% | 28.8% |
| SAGE 1.1 (Vals AI) | 85.4% | 88.7% |
| Chess Puzzles (Epoch AI) | 48% | 41% |
| SimpleQA Verified | 42.9% | 56% |
| Furniture Assembly (Epoch AI) | 34.2% | 20.8% |
| LiveBench Agentic Coding (LiveBench) | 77.3% | 54% |
| LiveBench Coding (LiveBench) | 80% | 77.2% |
| LiveBench Data Analysis (LiveBench) | 79.3% | 76.9% |
| LiveBench Instruction Following (LiveBench) | 70% | 75.3% |
| LiveBench Language (LiveBench) | 81.2% | 80.1% |
| LiveBench Mathematics (LiveBench) | 93.3% | 95.7% |
| LiveBench Reasoning (LiveBench) | 86.7% | 82.7% |
| SAGE (Vals AI) | 47.9% | 40.8% |
| DeepSWE v1.1 | 74.2% | 71% |
FAQ
Is DeepSeek V4.1 Flash better than Grok 4.7?
Grok 4.7 scores 53.0 on the Respan Index (#34), ahead of DeepSeek V4.1 Flash at 52.5 (#37). DeepSeek V4.1 Flash leads on 17 of the 25 benchmarks both report.
Which is cheaper, DeepSeek V4.1 Flash or Grok 4.7?
DeepSeek V4.1 Flash lists at $0.051 input and $0.203 output per 1M tokens; Grok 4.7 at $2 and $6. DeepSeek V4.1 Flash is 34x cheaper per token at list price.
Which has the larger context window?
DeepSeek V4.1 Flash accepts 1M tokens and Grok 4.7 500K tokens.