Inkling Thinking
Databricks · inkling
Multimodal MoE reasoning model (975B total, 41B active) for text, image, and audio
Specification
databricks listing
Context
1M
Max output
33K
Released
2026-07-15
Knowledge cutoff
—
Retires
—
Open weights
yes
Input
text, image, audio
Output
text
Price
US dollars per million tokens · databricks list price
Input
$1.00
Output
$4.05
Cache read
$0.17
Cache write
$1.00
Quality
7 benchmarks · 2 sources
Composite
61st
percentile of 334 scored models
Rank
132 / 334
±0.08 sd
Evidence
7 × 2
3 or more benchmarks, from 2 or more sources
Effort range
—
one configuration only
reasoning86th2/4 bench
math49th3/7 bench
knowledge45th1/1 bench
general27th1/1 bench
Every score
one row per source and configuration — nothing averaged away
| Benchmark | Score | Configuration | Source | Run |
|---|---|---|---|---|
| GPQA diamond | 88.3 ±1.9 | xhigh | Epoch AI | 2026-08-05 |
| OTIS Mock AIME 2024-2025 | 88.9 ±3.9 | xhigh | Epoch AI | 2026-08-05 |
| Chess Puzzles | 21.0 ±4.1 | xhigh | Epoch AI | 2026-08-05 |
| SimpleQA Verified | 40.3 ±1.6 | xhigh | Epoch AI | 2026-08-27 |
| FrontierMath-Tiers-1-3-v2 | 33.3 ±2.8 | xhigh | Epoch AI | 2026-08-06 |
| LiveBench | 72.9 | xhigh · 23 runs | LiveBench | 2026-06-25 |
| FrontierMath-Tier-4-v2 | 4.9 ±3.4 | xhigh | Epoch AI | 2026-08-06 |
The leading figure for a benchmark is the median across its configurations, so one heroic high-effort run cannot set the number.
Available from 26 hosts
| Host | In $/M | Out $/M | Cache rd | Cache wr | Context | Output | Retires |
|---|---|---|---|---|---|---|---|
| databricksvendor | $1.00 | $4.05 | $0.17 | $1.00 | 1M | — | — |
| nvidia · thinkingmachines | free | free | — | — | 1M | 16K | — |
| openrouter · free | free | free | — | — | 1M | 262K | — |
| deepinfra | $0.95 | $4.05 | $0.16 | — | 524K | — | — |
| deepinfra · thinkingmachines | $0.95 | $4.05 | $0.16 | — | 524K | 1M | — |
| kilo · thinkingmachines | $0.95 | $4.05 | $0.16 | — | 524K | 472K | — |
| baseten · thinkingmachines | $1.00 | $4.05 | — | — | 1M | 33K | — |
| edenai | $1.00 $0.95 | $4.05 | $0.17 $0.16 | $1.00 | 1M 524K 1M | 1M | — |
| fireworks_ai | $1.00 | $4.05 | $0.17 | — | 1M | — | — |
| fireworks-ai · accounts | $1.00 | $4.05 | $0.17 | — | 1M | 1M | — |
| huggingface · thinkingmachines | $1.00 | $4.05 | — | — | 1M | 1M | — |
| merge-gateway · thinkingmachines | $1.00 | $4.05 | $0.17 | — | 1M | 32K | — |
| nano-gpt · thinking | $1.00 | $4.05 | $0.17 | — | 1M | 33K | — |
| nano-gpt · thinkingmachines | $1.00 | $4.05 | $0.17 | — | 1M | 33K | — |
| neon | $1.00 | $4.05 | $0.17 | — | 1M | 66K | — |
| openrouter · thinkingmachines | $1.00 | $4.05 | $0.17 | — | 1M | 472K | — |
| together_ai | $1.00 | $4.05 | $0.17 | — | 524K | — | — |
| togetherai · thinkingmachines | $1.00 | $4.05 | $0.17 | — | 524K | 131K | — |
| vercel · thinkingmachines | $1.00 | $4.05 | $0.17 | — | 256K | 256K | — |
| hyper | $1.0888 | $4.4096 | $0.1851 | — | 1M | 33K | — |
| venice | $1.25 | $5.0625 | $0.2125 | — | 524K | 66K | — |
| impossibl · thinkingmachines | $1.87 | $4.68 | $0.374 | — | 66K | 66K | — |
| llmtr · thinkingmachines | $1.87 | $4.68 | — | — | 262K | 262K | — |
| requesty | $1.87 | $4.68 | $0.374 | — | 66K | 33K | — |
| thinkingmachines · thinkingmachines | $1.87 | $4.68 | $0.374 | — | 66K | 66K | — |
| abacus · thinkingmachines | $3.74 | $9.36 | $0.748 | — | 262K | 131K | — |
Price history
append-only observations · a listing writes a row only when its price moves
| Date | Host | Field | $/M |
|---|---|---|---|
| 2026-09-02 | edenai | Cache read | $0.16 |
| 2026-09-07 | edenai | Cache read | $0.17 |
| 2026-09-02 | edenai | Input | $0.95 |
| 2026-09-07 | edenai | Input | $1.00 |
More from Databricks
most-hosted first