Model comparison

Inkling vs Qwen3.5-122B-A10B

Data verified July 15, 2026

Head-to-head evidence from 5 shared benchmark results across 4 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.

Inkling

Thinking Machines Lab

69/100

Margin

8.5pts

← winning

Qwen3.5-122B-A10B

Alibaba

60.5/100

1 category wins4 category wins

Verified leaderboard positions: Inkling #17; Qwen3.5-122B-A10B #25

BenchAlign evidence: Inkling not scored; Qwen3.5-122B-A10B supported. Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.

Evidence parity. Inkling and Qwen3.5-122B-A10B share 5 comparable benchmark results. 5 of 8 categories are comparable. 11 results are unique to Inkling; 27 to Qwen3.5-122B-A10B.

Updated July 15, 2026

Shared results: 5
Inkling only: 11
Qwen3.5-122B-A10B only: 27
Comparable categories: 5 / 8

Pick Inkling if you want the stronger benchmark profile. Qwen3.5-122B-A10B only becomes the better choice if knowledge is the priority or you want the cheaper token bill.

Confidence note. This is a partial-evidence comparison with 5 shared benchmark results across 4 evidence categories; 5 of 8 categories currently have scoreable aggregates for both models. Treat the verdict as directional until coverage is more balanced.

Why this result

Inkling is clearly ahead on the provisional aggregate, 69 to 59. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.

Inkling's sharpest advantage is in agentic, where it averages 69.4 against 56.4. The single biggest benchmark swing on the page is Terminal-Bench 2.0, 63.8% to 49.4%. Qwen3.5-122B-A10B does hit back in knowledge, so the answer changes if that is the part of the workload you care about most.

Inkling is also the more expensive model on tokens at $1.87 input / $4.68 output per 1M tokens, versus $0.00 input / $0.00 output per 1M tokens for Qwen3.5-122B-A10B. That is roughly Infinityx on output cost alone. Qwen3.5-122B-A10B is the reasoning model in the pair, while Inkling is not. That usually helps on harder chain-of-thought-heavy tests, but it can also mean more latency and more token spend in real use. Inkling gives you the larger context window at 1M, compared with 262K for Qwen3.5-122B-A10B.

Category breakdown

Exact category averages are shown below. Not measured means BenchLM does not have enough sourced public coverage for that model and category.

Category scores and score margins for Inkling and Qwen3.5-122B-A10B
Category	Inkling	Δ	Qwen3.5-122B-A10B
Knowledge	Inkling51.7	Margin→ 31.9	Qwen3.5-122B-A10B83.6
Inst. Following	Inkling79.8	Margin→ 13.6	Qwen3.5-122B-A10B93.4
Agentic	Inkling69.4	Margin← 13.0	Qwen3.5-122B-A10B56.4
Coding	Inkling68.6	Margin→ 3.4	Qwen3.5-122B-A10B72.0
Multimodal	Inkling76.5	Margin→ 0.7	Qwen3.5-122B-A10B77.2
Reasoning	InklingNot measured	MarginNo overlap	Qwen3.5-122B-A10B60.2
Math	Inkling97.1	MarginNo overlap	Qwen3.5-122B-A10BNot measured
Multilingual	InklingNot measured	MarginNo overlap	Qwen3.5-122B-A10B82.2

Decisive benchmark drivers

The largest measured benchmark gaps in this matchup, with exact reported values.

A · InklingB · Qwen3.5-122B-A10B

Terminal-Bench 2.0
Agentic
Source ↗
A 63.8%B 49.4%
Winner: InklingΔ 14.4
Terminal-Bench 2.0: Inkling scored 63.8%; Qwen3.5-122B-A10B scored 49.4%. Inkling wins this benchmark.
BrowseComp
Agentic
Source ↗
A 77.1%B 63.8%
Winner: InklingΔ 13.3
BrowseComp: Inkling scored 77.1%; Qwen3.5-122B-A10B scored 63.8%. Inkling wins this benchmark.
SWE-bench Verified
Coding
Source ↗
A 77.6%B 72%
Winner: InklingΔ 5.6
SWE-bench Verified: Inkling scored 77.6%; Qwen3.5-122B-A10B scored 72%. Inkling wins this benchmark.
CharXiv
Multimodal
Source ↗
A 82%B 77.2%
Winner: InklingΔ 4.8
CharXiv: Inkling scored 82%; Qwen3.5-122B-A10B scored 77.2%. Inkling wins this benchmark.
GPQA
Knowledge
Source ↗
A 87.9%B 86.6%
Winner: InklingΔ 1.3
GPQA: Inkling scored 87.9%; Qwen3.5-122B-A10B scored 86.6%. Inkling wins this benchmark.

Operational comparison

Runtime and commercial metrics are compared only when both models have a complete sourced value.

Metric	Inkling	Qwen3.5-122B-A10B	Comparison
Input / output priceUSD per 1M tokens	Inkling$1.87 input / $4.68 output	Qwen3.5-122B-A10B$0 input / $0 output	Qwen3.5-122B-A10B has the lower combined listed price.
Generation speedtokens per second	InklingNot available	Qwen3.5-122B-A10BNot available	A complete speed comparison is not available.
First-answer latencyseconds to first token	InklingNot available	Qwen3.5-122B-A10BNot available	A complete latency comparison is not available.
Context windowmaximum listed tokens	Inkling1M	Qwen3.5-122B-A10B262K	Inkling lists the larger context window.

Benchmark Deep Dive

AgenticInkling wins

9 benchmarks

Benchmark	Inkling	Qwen3.5-122B-A10B	Result
Terminal-Bench 2.0Source	63.8%	49.4%	Inkling leads
BrowseCompSource	77.1%	63.8%	Inkling leads
MCP AtlasSource	74.1%	—	Not comparable
Design Arena Agentic Web DevSource	1258	—	Not comparable
OSWorld-VerifiedSource	—	58%	Not comparable
τ²-bench resultsSource	—	93.6%	Not comparable
AA Agentic IndexSource	—	20.7%	Not comparable
GDPval-AASource	—	23.9%	Not comparable
GDPval-AASource	—	978	Not comparable

CodingQwen3.5-122B-A10B wins

6 benchmarks

Benchmark	Inkling	Qwen3.5-122B-A10B	Result
SWE-bench VerifiedSource	77.6%	72%	Inkling leads
SWE-bench ProSource	54.3%	—	Not comparable
Terminal-Bench 2.0Source	63.8%	—	Not comparable
AA Coding IndexSource	—	45.7%	Not comparable
Terminal-Bench HardSource	—	31.1%	Not comparable
AA-SciCodeSource	—	42.0%	Not comparable

Reasoning

3 benchmarks

Benchmark	Inkling	Qwen3.5-122B-A10B	Result
LongBench v2Source	—	60.2%	Not comparable
AA-LCRSource	—	66.7%	Not comparable
CritPtSource	—	0.6%	Not comparable

KnowledgeQwen3.5-122B-A10B wins

12 benchmarks

Benchmark	Inkling	Qwen3.5-122B-A10B	Result
GPQASource	87.9%	86.6%	Inkling leads
GPQA-DSource	87.9%	—	Not comparable
HLESource	46%	—	Not comparable
HLE w/o toolsSource	30%	—	Not comparable
MMLU-ProSource	—	86.7%	Not comparable
SuperGPQASource	—	67.1%	Not comparable
Artificial Analysis Intelligence IndexSource	—	32.3%	Not comparable
AA-GPQA DiamondSource	—	85.7%	Not comparable
AA-HLESource	—	23.4%	Not comparable
AA-Omniscience IndexSource	—	-39.6%	Not comparable
AA-Omniscience AccuracySource	—	24.7%	Not comparable
AA-Omniscience Hallucination RateSource	—	85.5%	Not comparable

Math

1 benchmarks

Benchmark	Inkling	Qwen3.5-122B-A10B	Result
AIME26Source	97.1%	—	Not comparable

Multilingual

1 benchmarks

Benchmark	Inkling	Qwen3.5-122B-A10B	Result
MMLU-ProXSource	—	82.2%	Not comparable

MultimodalQwen3.5-122B-A10B wins

8 benchmarks

Benchmark	Inkling	Qwen3.5-122B-A10B	Result
MMMU-ProSource	73.5%	—	Not comparable
CharXivSource	82%	77.2%	Inkling leads
CharXiv w/o toolsSource	78.1%	—	Not comparable
MMMUSource	—	83.9%	Not comparable
MMVUSource	—	74.7%	Not comparable
MathVisionSource	—	86.2%	Not comparable
V*Source	—	93.2%	Not comparable
AA-MMMU-ProSource	—	75.0%	Not comparable

Inst. FollowingQwen3.5-122B-A10B wins

3 benchmarks

Benchmark	Inkling	Qwen3.5-122B-A10B	Result
IFBenchSource	79.8%	—	Not comparable
IFEvalSource	—	93.4%	Not comparable
AA-IFBenchSource	—	75.7%	Not comparable

Frequently Asked Questions (6)

Which is better, Inkling or Qwen3.5-122B-A10B?

Inkling is ahead on BenchLM's provisional leaderboard, 69 to 59. The biggest single separator in this matchup is Terminal-Bench 2.0, where the scores are 63.8% and 49.4%.

Which is better for knowledge tasks, Inkling or Qwen3.5-122B-A10B?

Qwen3.5-122B-A10B has the edge for knowledge tasks in this comparison, averaging 83.6 versus 51.7. Inside this category, GPQA is the benchmark that creates the most daylight between them.

Which is better for coding, Inkling or Qwen3.5-122B-A10B?

Qwen3.5-122B-A10B has the edge for coding in this comparison, averaging 72 versus 68.6. Inside this category, SWE-bench Verified is the benchmark that creates the most daylight between them.

Which is better for agentic tasks, Inkling or Qwen3.5-122B-A10B?

Inkling has the edge for agentic tasks in this comparison, averaging 69.4 versus 56.4. Inside this category, Terminal-Bench 2.0 is the benchmark that creates the most daylight between them.

Which is better for multimodal and grounded tasks, Inkling or Qwen3.5-122B-A10B?

Qwen3.5-122B-A10B has the edge for multimodal and grounded tasks in this comparison, averaging 77.2 versus 76.5. Inside this category, CharXiv is the benchmark that creates the most daylight between them.

Which is better for instruction following, Inkling or Qwen3.5-122B-A10B?

Qwen3.5-122B-A10B has the edge for instruction following in this comparison, averaging 93.4 versus 79.8. Inkling stays close enough that the answer can still flip depending on your workload.

Related Comparisons

Explore More

Thinking Machines Lab Compare Pricing Methodology Find Your Best LLM Overall Rankings

Last updated: July 15, 2026

The AI models change fast. We track them for you.

A weekly brief for engineers and researchers covering new models, ranking shifts, and pricing changes.

Free. No spam. Unsubscribe anytime.