DeepSeek V4.1 Flash 1017 vs GPT-5.6 Terra

Change models
IndexAll categories, out of 100.
68.5 ±2.6
68.9 ±2.0 best
GapIndex points below the top.
−0.4 · no clear gap
Top
CoverageIndex weight that has results.
95%
100% best
SpeedOutput tokens per second.
261/s best
66/s
EloLMArena rating and rank.
1463 best
1449
ReleasedMonth of release.
2026-10
2026-07
Input / 1MUS dollars per 1M input tokens.
$0.022 best
$2.00
Output / 1MUS dollars per 1M output tokens.
$0.40 best
$12.00
Batch output / 1MBatch price per 1M output tokens.
$0.20 best
$6.00
1K chats2K tokens in, 500 out, each.
$0.244 best
$10.00
1K documents20K tokens in, 1K out, each.
$0.84 best
$52.00
1K agent tasks200K tokens in, 20K out, each.
$12.40 best
$640.00
ContextMaximum tokens in one request.
1.05M
1.05M
AgenticMulti-step tasks with tools.
72.2 best
68.2
CodingCode writing and repair.
69.8 best
68.4
ReasoningLogic problems and puzzles.
65.0
72.5 best
MultimodalTasks with images and text.
62.2
67.3 best
KnowledgeFacts and expert knowledge.
64.2
65.5 best
MultilingualTasks in many languages.
—
60.7
InstructionTasks with strict prompt rules.
68.3 best
56.4
MathMath problems.
65.4
72.8 best

Benchmarks

69 shared
Only DeepSeek V4.1 Flash 1017: 25Only GPT-5.6 Terra: 40
DeepSeek V4.1 Flash 1017
GPT-5.6 Terra
39.2%51 / 222
42.9% best34 / 222
84.0% best8 / 217
83.0%18 / 217
89.8%42 / 217
93.3% best19 / 217
14.3%52 / 214
30.0% best10 / 214
98.3%28 / 200
99.7% best14 / 200
46.4%33 / 164
46.8% best31 / 164
83.3%56 / 149
85.1% best25 / 149
90.9%25 / 141
92.9% best12 / 141
55.0% best21 / 133
47.6%33 / 133
72.9%25 / 131
83.9% best17 / 131
94.5%21 / 126
96.5% best11 / 126
84.7% best18 / 123
74.6%44 / 123
81.7%23 / 122
91.3% best13 / 122
98.0%11 / 116
98.3% best8 / 116
51.9%44 / 114
55.0% best29 / 114
77.0%43 / 113
80.7% best17 / 113
MedScribeHealthcare
85.5% best27 / 109
82.9%50 / 109
MedCodeHealthcare
41.2%56 / 106
43.4% best45 / 106
42.9%47 / 91
43.2% best46 / 91
Chess PuzzlesReasoning
48.0% best19 / 90
47.0%24 / 90