Inkling-Small vs GPT-5

Change models
Inkling-SmallThinking Machines Lab
GPT-5OpenAI
IndexAll categories, out of 100.
55.1 ±3.2 best
55.0 ±4.2
GapIndex points below the top.
Top
−0.1 · no clear gap
CoverageIndex weight that has results.
95% best
90%
SpeedOutput tokens per second.
80/s best
76/s
EloLMArena rating and rank.
1414 best
1406
ReleasedMonth of release.
2026-07
2025-08
Input / 1MUS dollars per 1M input tokens.
$0.45 best
$1.25
Output / 1MUS dollars per 1M output tokens.
$1.20 best
$10.00
Batch output / 1MBatch price per 1M output tokens.
—
$5.00
1K chats2K tokens in, 500 out, each.
$1.50 best
$7.50
1K documents20K tokens in, 1K out, each.
$10.20 best
$35.00
1K agent tasks200K tokens in, 20K out, each.
$114.00 best
$450.00
ContextMaximum tokens in one request.
524K best
400K
AgenticMulti-step tasks with tools.
55.7 best
53.1
CodingCode writing and repair.
54.9 best
52.5
ReasoningLogic problems and puzzles.
55.7 best
55.4
MultimodalTasks with images and text.
54.7
61.2 best
KnowledgeFacts and expert knowledge.
52.0
58.4 best
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
65.8
—
MathMath problems.
55.3
55.3

Benchmarks

21 shared
Only Inkling-Small: 43Only GPT-5: 24
Inkling-Small
GPT-5
88.5% best53 / 217
86.2%68 / 217
90.0%62 / 200
91.4% best56 / 200
83.0%62 / 149
86.0% best17 / 149
TaxEval v2Finance
75.5% best18 / 143
73.4%49 / 143
85.9%25 / 142
85.9%25 / 142
83.6%55 / 135
85.6% best45 / 135
MMLU ProKnowledge
85.6%55 / 135
86.5% best41 / 135
40.1% best53 / 131
9.9%68 / 131
CorpFin v2Finance
69.6% best6 / 131
61.1%58 / 131
84.0% best52 / 126
65.7%62 / 126
19.1%95 / 123
20.1% best91 / 123
39.3% best46 / 122
9.6%59 / 122
89.4% best47 / 116
65.9%62 / 116
MedScribeHealthcare
84.1% best38 / 109
83.7%42 / 109
MedCodeHealthcare
37.9%74 / 106
49.6% best21 / 106
62.3%55 / 97
65.5% best41 / 97
19.1%79 / 91
50.1% best30 / 91
46.3%57 / 90
55.4% best51 / 90
SWE-benchCoding
82.2% best20 / 87
69.0%67 / 87
6.0%74 / 80
23.0% best39 / 80