Inkling (xhigh): benchmark scores, price and ranking
Inkling (xhigh) from Thinking Machines ranks #314 of 743 LLMs on the LLMs Tiger aggregate benchmark score (48.9), based on 11 benchmark results captured up to 2026-10-13.
- Organisation
- Thinking Machines
- Type
- Open-weight / local model
- Open weights
- Yes
- Released
- 2026-07-15 (source: epoch)
- Cheapest API price
- $0.950 in / $4.05 out per million tokens (via deepinfra, observed 2026-10-13T19:47Z)
- Aggregate rank
- #314 of 743 (score 48.9)
Relative to other models, Inkling (xhigh) is strongest on GPQA Diamond (#49 of 275) and weakest on FrontierSWE (#19 of 19).
Across the 11 benchmarks where at least ten models are scored, Inkling (xhigh) is in the top 10% on 0 and in the top half on 7.
Its cheapest listed input price, $0.950 per million tokens, is lower than that of 37% of the 427 models with a listed price.
Inkling (xhigh) benchmark results
| Benchmark | Score | Rank | Source | Captured |
|---|---|---|---|---|
| GPQA Diamond | 88.3% | #49 of 275 | Epoch AI Benchmarking Hub | 2026-10-13 |
| DTBench | 87.5% | #33 of 146 | Epoch AI Benchmarking Hub | 2026-10-13 |
| OTIS Mock AIME | 88.9% | #47 of 189 | Epoch AI Benchmarking Hub | 2026-10-13 |
| LMCA | 37.6% | #35 of 108 | Epoch AI Benchmarking Hub | 2026-10-13 |
| Chess Puzzles | 21.0% | #41 of 122 | Epoch AI Benchmarking Hub | 2026-10-13 |
| CritPt | 5.4% | #47 of 115 | Epoch AI Benchmarking Hub | 2026-10-13 |
| SciCode | 47.0% | #45 of 106 | Epoch AI Benchmarking Hub | 2026-10-13 |
| SimpleQA Verified | 40.3% | #38 of 69 | Epoch AI Benchmarking Hub | 2026-10-13 |
| FrontierMath T1-3 | 33.3% | #49 of 68 | Epoch AI Benchmarking Hub | 2026-10-13 |
| FrontierMath T4 | 4.9% | #44 of 50 | Epoch AI Benchmarking Hub | 2026-10-13 |
| FrontierSWE | 4.1% | #19 of 19 | Epoch AI Benchmarking Hub | 2026-10-13 |
About Inkling (xhigh)
Inkling (xhigh) is an open-weight model that can run on your own hardware from Thinking Machines. It was released on 2026-07-15, according to Epoch AI. The cheapest API listing we track charges $0.950 per million input tokens and $4.05 per million output tokens, through deepinfra. We hold 11 benchmark results for it across 4 areas, and none is in the top tenth of its board.
Where Inkling (xhigh) stands
On our aggregate score, Inkling (xhigh) is #314 of 743 ranked models, around the middle of the table. Its score is 48.9; the scale is centred on 50, so it sits 1.1 points below the typical model. That beats 58% of the ranked field. Among the 2 ranked Thinking Machines models it is #2. The next one up is Inkling Small (xhigh) at 49.7.
Inkling (xhigh), benchmark by benchmark
GPQA Diamond puts Inkling (xhigh) at 88.3%, #49 of 275. That beats 82% of the models tested, but 7.5 points short of GPT-6 Astra (max) (95.8%). The results either side are Ring-2.6-1T (88.3%) and NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4 (87.9%). It covers expert-level science questions. The board median is 71.7%, so Inkling (xhigh) is 16.5 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
On DTBench, Inkling (xhigh) reaches 87.5%. It ranks #33 of 146, ahead of 77% of the field, and 11.5 points behind Claude Opus 5.5 (max) at 98.9%. The results either side are grok-4-1-fast-reasoning (87.7%) and Qwen3.5 397B-A17B (87.5%). It covers dynamic tool-use benchmark. The board median is 68.8%, so Inkling (xhigh) is 18.7 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
Inkling (xhigh) manages 88.9% on OTIS Mock AIME, good for #47 of 189. Claude Fable 5.1 (max) leads at 100.0%, 11.1 points further up, and 75% of models sit below. The results either side are gpt-oss-120b (high) (88.9%) and Qwen3.5 397B-A17B (88.9%). It covers competition maths (AIME 2024-25 style). The board median is 57.8%, so Inkling (xhigh) is 31.1 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
LMCA puts Inkling (xhigh) at 37.6%, #35 of 108. That beats 68% of the models tested, but 30.7 points short of Claude Opus 5.5 (max) (68.2%). The results either side are Qwen3.5 397B-A17B (37.9%) and Qwen3.7 Plus (37.6%). It covers language-model capability assessment. The board median is 29.2%, so Inkling (xhigh) is 8.4 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
On Chess Puzzles, Inkling (xhigh) reaches 21.0%. It ranks #41 of 122, ahead of 66% of the field, and 51.0 points behind GPT-6 Astra (max) at 72.0%. The results either side are GLM-5.3 (max) (21.0%) and Kimi K2.7 Code (21.0%). It covers chess tactics puzzles. The board median is 12.5%, so Inkling (xhigh) is 8.5 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
Inkling (xhigh) manages 5.4% on CritPt, good for #47 of 115. GPT-5.6 Sol (max) leads at 32.3%, 26.9 points further up, and 59% of models sit below. The results either side are Gemini 3 Pro Preview (6.9%) and Qwen3.8-27B (XHigh) (5.4%). It covers unpublished research-level physics problems. The board median is 2.9%, so Inkling (xhigh) is 2.6 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
Inkling (xhigh) lands below the middle on SimpleQA Verified: 40.3%, #38 of 69, with 54% of the field scoring higher. gemini-4-argon-high leads at 76.0%; the closest result above is Qwen3-235B-A22B-Thinking-2507 at 40.4%. It covers short factual questions. The board median is 41.0%, so Inkling (xhigh) is 0.7 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
Inkling (xhigh) lands below the middle on FrontierMath T1-3: 33.3%, #49 of 68, with 71% of the field scoring higher. GPT-6 Astra (max) leads at 93.7%; the closest result above is Qwen3.7 Plus at 34.4%. It covers research-level maths, tiers 1-3. The board median is 55.6%, so Inkling (xhigh) is 22.3 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
FrontierMath T4 is a weak spot. Inkling (xhigh) scores 4.9%, #44 of 50, behind 86% of the models measured. The leader, GPT-6.1 Sol (max), reaches 100.0%. It covers research-level maths, tier 4. The board median is 30.5%, so Inkling (xhigh) is 25.6 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
FrontierSWE is a weak spot. Inkling (xhigh) scores 4.1%, #19 of 19, behind 95% of the models measured. The leader, GPT-6 Astra (max), reaches 65.5%. It covers hard multi-hour software engineering. The board median is 29.5%, so Inkling (xhigh) is 25.4 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
The full table above also lists the other 1 result, including boards with fewer than ten models, where a rank says little.
Strengths and weaknesses
Software engineering: weak, ahead of 29% of the field on average across 2 benchmarks. Its best board here is SciCode (#45 of 106) and its weakest is FrontierSWE (#19 of 19). Reasoning, maths and science: solid, ahead of 51% of the field on average across 5 benchmarks. Its best board here is GPQA Diamond (#49 of 275) and its weakest is FrontierMath T4 (#44 of 50). General knowledge and chat: solid, ahead of 60% of the field on average across 3 benchmarks. Its best board here is LMCA (#35 of 108) and its weakest is SimpleQA Verified (#38 of 69).
Price and value
The median input price across the 427 priced models is $0.500 per million tokens, so Inkling (xhigh) costs more than the typical listing. Models that score within 1.5 points or better and cost less: Qwen2.5-Coder-7B-Instruct ($0.010, 49.5), DeepSeek V4.1 Flash (high) ($0.010, 48.9), Qwen3.7 Flash ($0.030, 50.5). Similar scores cost more on GLM-5.1 ($0.966) and Claude 4.5 Haiku ($1.00).
Release history
The Thinking Machines model released before it was Inkling Small (xhigh) on 2026-07-15, which scores 49.7.
Reading these numbers
The 11 scores on this page come from 1 source: Epoch AI Benchmarking Hub (11). No official leaderboard supplied any of them, so treat them as third-party measurements. The leaderboard also lists 2 other reasoning-effort settings of this model, which redirect here. This page shows the highest.
Inkling (xhigh) is also listed at 2 other reasoning-effort settings; this page shows the highest-effort row, which is the one the leaderboard keeps by default.
Models ranked nearby
- #311 Yi Coder 9B Chat (48.9)
- #312 deepseek-v4-pro-high-20260813 (48.9)
- #313 qwen-plus-2025-04-28 (48.9)
- #315 grok-4.3 (high) (48.9)
- #316 claude-opus-4-5-20251101-high-32k (48.9)
- #317 DeepSeek V4.1 Flash (high) (48.9)
More from Thinking Machines
- Inkling Small (xhigh) (#244, 49.7)
Compare it on the live leaderboard → · How the aggregate score works