Capable on Reasoning and Agentic +3 more; limited on Coding and Math +1 more — at $0.95/M input across 5 providers. Insufficient data on 2 of 10 axes.
| Provider | Model name | Input $/M | Output $/M | Effective | Current |
|---|---|---|---|---|---|
| deepinfra ↗ | thinkingmachines/Inkling | $0.95 | $4.05 | 2026-09-02 | ● |
| direct ↗ | inkling | $1.00 | $4.05 | 2026-09-02 | ● |
| thinkingmachines ↗ | inkling | $1.00 | $4.05 | 2026-09-02 | ● |
| baseten ↗ | thinkingmachines/Inkling | $1.00 | $4.05 | 2026-09-01 | ● |
| together ↗ | thinkingmachines/Inkling | $1.00 | $4.05 | 2026-09-01 | ● |
| Benchmark | Score | Settings | Source | Verification | Methodology | Measured | Reference |
|---|---|---|---|---|---|---|---|
livebench_instruction_following@2026-06-25 | 70.10 | x-high | livebench.ai | VERIFIED | OTHER | 2h ago | view ↗ |
livebench_reasoning@2026-06-25 | 78.35 | x-high | livebench.ai | VERIFIED | OTHER | 2h ago | view ↗ |
livebench_data_analysis@2026-06-25 | 72.78 | x-high | livebench.ai | VERIFIED | OTHER | 2h ago | view ↗ |
livebench_coding@2026-06-25 | 71.02 | x-high | livebench.ai | VERIFIED | OTHER | 2h ago | view ↗ |
livebench_agentic_coding@2026-06-25 | 49.39 | x-high | livebench.ai | VERIFIED | OTHER | 2h ago | view ↗ |
livebench_math@2026-06-25 | 88.36 | x-high | livebench.ai | VERIFIED | OTHER | 2h ago | view ↗ |
livebench_language@2026-06-25 | 73.46 | x-high | livebench.ai | VERIFIED | OTHER | 2h ago | view ↗ |
omniscienceAccuracy | 41.55 | x-high | artificialanalysis.ai | AGGREGATOR | AGGR | 3h ago | view ↗ |
aa_intelligence_index | 42.29 | x-high | artificialanalysis.ai | AGGREGATOR | AGGR | 3h ago | view ↗ |
aa_lcr | 73.33 | x-high | artificialanalysis.ai | AGGREGATOR | AGGR | 3h ago | view ↗ |
aa_omniscience | 2.00 | x-high | artificialanalysis.ai | AGGREGATOR | AGGR | 3h ago | view ↗ |
aa_coding_index | 52.06 | x-high | artificialanalysis.ai | AGGREGATOR | AGGR | 3h ago | view ↗ |
terminalbenchV21 | 55.06 | x-high | artificialanalysis.ai | AGGREGATOR | AGGR | 3h ago | view ↗ |
| 5.43 | x-high | artificialanalysis.ai | AGGREGATOR | AGGR | 3h ago | view ↗ | |
tauBanking | 29.07 | x-high | artificialanalysis.ai | AGGREGATOR | AGGR | 3h ago | view ↗ |
| 36.71 | x-high | artificialanalysis.ai | AGGREGATOR | AGGR | 3h ago | view ↗ | |
gpqa | 87.17 | x-high | artificialanalysis.ai | AGGREGATOR | AGGR | 3h ago | view ↗ |
aa_hle | 31.88 | x-high | artificialanalysis.ai | AGGREGATOR | AGGR | 3h ago | view ↗ |
aa_scicode | 46.06 | x-high | artificialanalysis.ai | AGGREGATOR | AGGR | 3h ago | view ↗ |
aa_agentic_index | 34.13 | x-high | artificialanalysis.ai | AGGREGATOR | AGGR | 3h ago | view ↗ |
omniscienceNonHallucination | 32.34 | x-high | artificialanalysis.ai | AGGREGATOR | AGGR | 3h ago | view ↗ |
aa_mmmu_pro | 73.47 | x-high | artificialanalysis.ai | AGGREGATOR | AGGR | 3h ago | view ↗ |
| 76.00 | x-high | labs.scale.commcp_atlas | VERIFIED | OTHER | 3h ago | view ↗ | |
| 1407.87 | — | datasets-server.huggingface.coleaderboard-dataset | VERIFIED | OTHER | 4d ago | view ↗ | |
| 40.30 | — | epoch.aibenchmarks.csv | VERIFIED | OTHER | 19d ago | view ↗ | |
ARC-AGI-1 | 79.50 | — | huggingface.coREADME | VENDOR | VENDOR | 33d ago | view ↗ |
HMMT Feb 2026 | 86.30 | — | huggingface.coREADME | VENDOR | VENDOR | 33d ago | view ↗ |
AIME 2026 | 97.10 | — | huggingface.coREADME | VENDOR | VENDOR | 33d ago | view ↗ |
| 46.00 | — | huggingface.coREADME | VENDOR | VENDOR | 33d ago | view ↗ | |
| 45.50 | — | huggingface.coREADME | VENDOR | VENDOR | 33d ago | view ↗ | |
BrowseComp (with context management) | 77.10 | — | huggingface.coREADME | VENDOR | VENDOR | 33d ago | view ↗ |
| 23.70 | — | huggingface.coREADME | VENDOR | VENDOR | 33d ago | view ↗ | |
| 1238.00 | x-high | huggingface.coREADME | VENDOR | VENDOR | 33d ago | view ↗ | |
SWEBench Pro (public) | 54.30 | — | huggingface.coREADME | VENDOR | VENDOR | 33d ago | view ↗ |
SWEBench Verified | 77.60 | — | huggingface.coREADME | VENDOR | VENDOR | 33d ago | view ↗ |
| 50.00 | — | simple-bench.comleaderboard-data.js | VERIFIED | OTHER | 41d ago | view ↗ | |
| 1445 | — | lmarena.ai | VERIFIED | AGGR | 41d ago | view ↗ | |
| 61.10 | — | thinkingmachines.aiintroducing-inkling | VENDOR | OTHER | 43d ago | view ↗ | |
| 36.53 | — | arcprize.orgdata | VERIFIED | OTHER | 46d ago | view ↗ | |
| 0.16 | x-high | thinkingmachines.aiintroducing-inkling | VENDOR | OTHER | 47d ago | view ↗ | |
StrongREJECT | 98.80 | x-high | thinkingmachines.aiintroducing-inkling | VENDOR | OTHER | 47d ago | view ↗ |
| 90.00 | x-high | thinkingmachines.aiintroducing-inkling | VENDOR | OTHER | 47d ago | view ↗ | |
| 77.50 | x-high | thinkingmachines.aiintroducing-inkling | VENDOR | OTHER | 47d ago | view ↗ | |
Charxiv RQ with python | 83.40 | x-high | thinkingmachines.aiintroducing-inkling | VENDOR | OTHER | 47d ago | view ↗ |
| 73.10 | x-high | thinkingmachines.aiintroducing-inkling | VENDOR | OTHER | 47d ago | view ↗ | |
| 86.80 | x-high | thinkingmachines.aiintroducing-inkling | VENDOR | OTHER | 47d ago | view ↗ | |
| 83.40 | x-high | thinkingmachines.aiintroducing-inkling | VENDOR | OTHER | 47d ago | view ↗ | |
| 77.10 | x-high | thinkingmachines.aiintroducing-inkling | VENDOR | OTHER | 47d ago | view ↗ | |
| 77.10 | — | huggingface.coREADME | VENDOR | VENDOR | 47d ago | view ↗ | |
CritPt | 5.40 | — | huggingface.coREADME | VENDOR | VENDOR | 33d ago | view ↗ |
HLE (text only) | 29.70 | — | huggingface.coREADME | VENDOR | VENDOR | 33d ago | view ↗ |
| 87.20 | — | huggingface.coREADME | VENDOR | VENDOR | 33d ago | view ↗ | |
SciCode | 46.10 | — | huggingface.coREADME | VENDOR | VENDOR | 33d ago | view ↗ |
| 79.50 | — | arcprize.orgdata | VERIFIED | OTHER | 46d ago | view ↗ | |
| 1407.60 | x-high | datasets-server.huggingface.coleaderboard-dataset | VERIFIED | OTHER | 47d ago | view ↗ | |
Artificial Analysis Intelligence Index | 42.00 | — | artificialanalysis.aiinkling | VERIFIED | OTHER | 47d ago | view ↗ |
SimpleQA Verified | 20.90 | x-high | thinkingmachines.aiintroducing-inkling | VENDOR | OTHER | 47d ago | view ↗ |
| 13.60 | x-high | thinkingmachines.aiintroducing-inkling | VENDOR | OTHER | 47d ago | view ↗ | |
MCP-Atlas | 74.90 | x-high | thinkingmachines.aiintroducing-inkling | VENDOR | OTHER | 47d ago | view ↗ |
| 53.20 | x-high | thinkingmachines.aiintroducing-inkling | VENDOR | OTHER | 47d ago | view ↗ | |
SWEBench Verified | 77.40 | x-high | thinkingmachines.aiintroducing-inkling | VENDOR | OTHER | 47d ago | view ↗ |
AIME 2026 | 95.10 | x-high | thinkingmachines.aiintroducing-inkling | VENDOR | OTHER | 47d ago | view ↗ |
HLE with tools | 46.60 | x-high | thinkingmachines.aiintroducing-inkling | VENDOR | OTHER | 47d ago | view ↗ |
Terminal Bench 2.1 | 63.80 | x-high | thinkingmachines.aiintroducing-inkling | VENDOR | OTHER | 47d ago | view ↗ |
StrongREJECT | 98.60 | — | huggingface.coREADME | VENDOR | VENDOR | 47d ago | view ↗ |
| 91.40 | — | huggingface.coREADME | VENDOR | VENDOR | 47d ago | view ↗ | |
| 77.20 | — | huggingface.coREADME | VENDOR | VENDOR | 47d ago | view ↗ | |
Charxiv RQ (with python) | 82.00 | — | huggingface.coREADME | VENDOR | VENDOR | 47d ago | view ↗ |
MMMU Pro (Standard 10) | 73.50 | — | huggingface.coREADME | VENDOR | VENDOR | 47d ago | view ↗ |
| 88.70 | — | huggingface.coREADME | VENDOR | VENDOR | 47d ago | view ↗ | |
| 79.80 | — | huggingface.coREADME | VENDOR | VENDOR | 47d ago | view ↗ | |
| 88.30 | x-high | thinkingmachines.aiintroducing-inkling | VENDOR | OTHER | 47d ago | view ↗ | |
HLE text only | 29.60 | x-high | thinkingmachines.aiintroducing-inkling | VENDOR | OTHER | 47d ago | view ↗ |
SimpleQA Verified | 43.90 | — | huggingface.coREADME | VENDOR | VENDOR | 47d ago | view ↗ |
| 74.10 | — | huggingface.coREADME | VENDOR | VENDOR | 47d ago | view ↗ | |
GDPVal-AA v2 | 1233.00 | — | huggingface.coREADME | VENDOR | VENDOR | 47d ago | view ↗ |