Inference
EN

DeepSeek V4.1 Flash 1017 vs GPT-6.1 Terra

Change models
DeepSeek V4.1 Flash 1017DeepSeekFastest
GPT-6.1 TerraOpenAI
IndexAll categories, out of 100.
68.5 ±2.6
77.1 ±3.8 best
GapIndex points below the top.
−8.6
Top
CoverageIndex weight that has results.
95% best
90%
SpeedOutput tokens per second.
261/s best
46/s
EloLMArena rating and rank.
1463
—
ReleasedMonth of release.
2026-10
2026-10
Input / 1MUS dollars per 1M input tokens.
$0.022 best
$1.00
Output / 1MUS dollars per 1M output tokens.
$0.40 best
$5.00
Batch output / 1MBatch price per 1M output tokens.
$0.20 best
$2.50
1K chats2K tokens in, 500 out, each.
$0.244 best
$4.50
1K documents20K tokens in, 1K out, each.
$0.84 best
$25.00
1K agent tasks200K tokens in, 20K out, each.
$12.40 best
$300.00
ContextMaximum tokens in one request.
1.05M
1.05M
AgenticMulti-step tasks with tools.
72.2
78.6 best
CodingCode writing and repair.
69.8
74.4 best
ReasoningLogic problems and puzzles.
65.0
81.3 best
MultimodalTasks with images and text.
62.2
73.2 best
KnowledgeFacts and expert knowledge.
64.2
70.5 best
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
68.3
—
MathMath problems.
65.4
76.9 best

Benchmarks

31 shared
Only DeepSeek V4.1 Flash 1017: 63Only GPT-6.1 Terra: 19
DeepSeek V4.1 Flash 1017
GPT-6.1 Terra
Artificial Analysis Humanity's Last ExamKnowledge
39.2%51 / 222
46.2% best24 / 222
Artificial Analysis Long Context ReasoningReasoning
84.0% best8 / 217
82.7%23 / 217
GPQA diamond · Epoch AIKnowledge
89.8%42 / 217
94.3% best10 / 217
Critical Physics TasksReasoning
14.3%52 / 214
28.0% best16 / 214
OTIS Mock AIME 2024-2025Math
98.3%28 / 200
100.0% best1 / 200
Artificial Analysis Omniscience AccuracyKnowledge
46.4%33 / 164
47.6% best28 / 164
Vibe Code Bench v1.1Coding
84.7% best18 / 123
84.0%19 / 123
Artificial Analysis SciCodeCoding
51.9%44 / 114
55.1% best28 / 114
Artificial Analysis MMMU-ProMultimodal
77.0%43 / 113
84.7% best6 / 113
SimpleQA VerifiedKnowledge
42.9%47 / 91
57.3% best21 / 91
FrontierMath-Tiers-1-3-v2-PrivateMath
67.4%33 / 90
90.9% best5 / 90
Legal Research BenchLegal
41.3% best30 / 90
26.0%62 / 90
Tax Agent BenchFinance
26.6% best41 / 82
17.1%58 / 82
LiveBench Data AnalysisReasoning
79.3%15 / 80
82.3% best4 / 80
LiveBench ReasoningReasoning
86.7%31 / 80
88.7% best20 / 80
Mystery Game PuzzlesReasoning
43.0%17 / 80
74.0% best4 / 80
FrontierMath-Tier-4-v2-PrivateMath
26.8%39 / 72
92.7% best6 / 72
Terminal-Bench 4.0Agentic
19.7%34 / 61
49.5% best11 / 61
Agent Arena steerabilityAgentic
4.018 / 54
16.6 best4 / 54
Agent Arena task outcomeAgentic
2.325 / 54
9.7 best11 / 54