# Inkling-Small vs Llama 3.1 70B

> Side-by-side benchmark comparison for Inkling-Small and Llama 3.1 70B across agentic, coding, multimodal, knowledge, reasoning, multilingual, and math tasks.

- Canonical page: https://benchlm.ai/compare/inkling-small-vs-llama-3-1-70b
- Last updated: October 14, 2026

- Shared sourced benchmarks: 1
- HTML indexing: indexable
- Ranking lane: BenchAlign v5.8

## Quick Verdict

Evidence is too thin for a benchmark verdict: Inkling-Small and Llama 3.1 70B share only 1 sourced benchmark row.

## Summary

- Only 1 shared sourced benchmark row is available. That is not enough for a broad capability verdict, so this mirror does not name an overall winner.
- Inkling-Small has the larger published context window at 1M.

## Model Snapshot

| Property | Inkling-Small | Llama 3.1 70B |
|----------|----------|----------|
| Creator | Thinking Machines Lab | Meta |
| Type | Open Weight | Open Weight |
| Reasoning | Hybrid | Non-Reasoning |
| Context | 1M | 128K |
| Independent public score (not head-to-head) | 53.8 | null |
| Benchmarks Covered | 28 | 7 |
| Pricing (input/output) | $0.58 / $1.44 | N/A |

## Category Breakdown

### Agentic

- Winner: Insufficient shared evidence
- Inkling-Small public-lane score: 37.1 (Supported · #74/124)
- Llama 3.1 70B public-lane score: Coming soon

| Benchmark | Inkling-Small | Llama 3.1 70B | Winner |
|-----------|-----------|-----------|--------|
| Terminal-Bench 2.1 | 64.7% | Coming soon | Coming soon |
| BrowseComp | 77.4% | Coming soon | Coming soon |
| MCP Atlas | 79.6% | Coming soon | Coming soon |
| Toolathlon-Verified | 54.4% | Coming soon | Coming soon |
| Terminal-Bench 2.1 (Vals) | 55.1% | Coming soon | Coming soon |

### Coding

- Winner: Insufficient shared evidence
- Inkling-Small public-lane score: 37.9 (Supported · #69/148)
- Llama 3.1 70B public-lane score: Coming soon

| Benchmark | Inkling-Small | Llama 3.1 70B | Winner |
|-----------|-----------|-----------|--------|
| SWE-bench Verified | 80.2% | Coming soon | Coming soon |
| SWE-bench Pro | 55.9% | Coming soon | Coming soon |
| Terminal-Bench 2.1 | 64.7% | Coming soon | Coming soon |
| SciCode | 48.7% | Coming soon | Coming soon |
| LiveCodeBench (Vals) | 85.9% | Coming soon | Coming soon |
| SWE-bench (Vals) | 82.2% | Coming soon | Coming soon |
| HumanEval | Coming soon | 80.5% | Coming soon |

### Reasoning

- Winner: Insufficient shared evidence
- Inkling-Small public-lane score: 55 (Unranked · 4 rankable rows)
- Llama 3.1 70B public-lane score: Coming soon

| Benchmark | Inkling-Small | Llama 3.1 70B | Winner |
|-----------|-----------|-----------|--------|
| ARC-AGI-2 | 40.1% | Coming soon | Coming soon |
| CritPt | 8.3% | Coming soon | Coming soon |
| ARC-AGI-1 | 84.00% | Coming soon | Coming soon |

### Multimodal & Grounded

- Winner: Insufficient shared evidence
- Inkling-Small public-lane score: 67.5 (#36/54)
- Llama 3.1 70B public-lane score: Coming soon

| Benchmark | Inkling-Small | Llama 3.1 70B | Winner |
|-----------|-----------|-----------|--------|
| MMMU-Pro | 74% | Coming soon | Coming soon |
| CharXiv | 81.3% | Coming soon | Coming soon |
| CharXiv w/o tools | 77.4% | Coming soon | Coming soon |

### Knowledge

- Winner: Insufficient shared evidence
- Inkling-Small public-lane score: 55.1 (Supported · #59/181)
- Llama 3.1 70B public-lane score: Coming soon

| Benchmark | Inkling-Small | Llama 3.1 70B | Winner |
|-----------|-----------|-----------|--------|
| GPQA | 89.5% | 46.7% | Inkling-Small |
| GPQA-D | 89.5% | Coming soon | Coming soon |
| HLE | 47.8% | Coming soon | Coming soon |
| HLE w/o tools | 31.6% | Coming soon | Coming soon |
| GPQA Diamond (Vals) | 83.6% | Coming soon | Coming soon |
| MMLU-Pro (Vals) | 85.6% | Coming soon | Coming soon |
| MMLU | Coming soon | 86% | Coming soon |
| MMLU-Pro | Coming soon | 66.4% | Coming soon |

### Multilingual

- Winner: Insufficient shared evidence
- Inkling-Small public-lane score: Coming soon
- Llama 3.1 70B public-lane score: Coming soon

| Benchmark | Inkling-Small | Llama 3.1 70B | Winner |
|-----------|-----------|-----------|--------|
| MGSM | Coming soon | 86.9% | Coming soon |

### Instruction Following

- Winner: Insufficient shared evidence
- Inkling-Small public-lane score: 93.3 (#9/128)
- Llama 3.1 70B public-lane score: Coming soon

| Benchmark | Inkling-Small | Llama 3.1 70B | Winner |
|-----------|-----------|-----------|--------|
| IFBench | 82.2% | Coming soon | Coming soon |
| IFEval | Coming soon | 87.5% | Coming soon |

### Mathematics

- Winner: Insufficient shared evidence
- Inkling-Small public-lane score: 69.8 (#5/12)
- Llama 3.1 70B public-lane score: Coming soon

| Benchmark | Inkling-Small | Llama 3.1 70B | Winner |
|-----------|-----------|-----------|--------|
| AIME26 | 95.5% | Coming soon | Coming soon |
| HMMT Feb 2026 | 90.2% | Coming soon | Coming soon |
| FrontierMath v2 (Tiers 1-3) | 46.316% | Coming soon | Coming soon |
| FrontierMath v2 (Tier 4) | 17.073% | Coming soon | Coming soon |
| GSM8K | Coming soon | 95.1% | Coming soon |

## FAQ

### Is there enough evidence to pick between Inkling-Small and Llama 3.1 70B?

Not yet. The models share only 1 sourced benchmark row, so this mirror reports the available measurements and metadata without naming a broad winner.

### What can I compare safely on this page?

Published pricing, context windows, model type, and individual shared benchmark rows are directly comparable. Independent overall scores and non-overlapping benchmark coverage are labeled separately and should not be read as a controlled head-to-head result.

## Related Comparisons

- [Inkling-Small vs Inkling](/compare/inkling-vs-inkling-small)
- [Llama 3.1 70B vs Inkling](/compare/inkling-vs-llama-3-1-70b)
- [Inkling-Small vs Claude Opus 5.5](/compare/claude-opus-5-5-vs-inkling-small)
- [Llama 3.1 70B vs Claude Opus 5.5](/compare/claude-opus-5-5-vs-llama-3-1-70b)

## Explore More

- [Inkling-Small profile](/models/inkling-small)
- [Llama 3.1 70B profile](/models/llama-3-1-70b)
- [Compare Pricing](/llm-pricing)
- [Alternative Finder](/tools/alternative-finder)
- [LLM Selector](/tools/llm-selector)
- [Overall Rankings](/best/overall)