Skip to main content

Model comparison

Celeris-1 vs Qwen3.5-122B-A10B

Data verified

Head-to-head evidence from 1 shared benchmark result across 1 category. Overall scores shown here use the public BenchAlign v5 ranking lane.

Celeris
N/A
No comparison
60.56/100
0 category wins1 category wins

Public leaderboard positions: Celeris-1 unranked (Not scored); Qwen3.5-122B-A10B #47 (Supported). Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.

Evidence parity. Celeris-1 and Qwen3.5-122B-A10B share 1 comparable benchmark result. 1 of 8 categories are comparable. 0 results are unique to Celeris-1; 30 to Qwen3.5-122B-A10B.

Updated July 24, 2026
Shared results
1
Celeris-1 only
0
Qwen3.5-122B-A10B only
30
Comparable categories
1 / 8

Treat this as a split decision. Celeris-1 makes more sense if you would rather avoid the extra latency and token burn of a reasoning model; Qwen3.5-122B-A10B is the better fit if knowledge is the priority or you want the cheaper token bill.

Confidence note. This is a partial-evidence comparison with 1 shared benchmark result across 1 evidence category; 1 of 8 categories currently have scoreable aggregates for both models. Treat the verdict as directional until coverage is more balanced.

Why this result

Celeris-1 and Qwen3.5-122B-A10B finish on the same BenchAlign overall score, so this is less about a single winner and more about where the edge shows up. The BenchAlign headline says tie; the benchmark table is where the real choice happens.

Celeris-1 is also the more expensive model on tokens at $2.00 input / $6.00 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 Celeris-1 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. Qwen3.5-122B-A10B gives you the larger context window at 262K, compared with 8K for Celeris-1.

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 Celeris-1 and Qwen3.5-122B-A10B
CategoryCeleris-1ΔQwen3.5-122B-A10B
KnowledgeCeleris-175.9Margin 7.7Qwen3.5-122B-A10B83.6
AgenticCeleris-1Not measuredMarginNo overlapQwen3.5-122B-A10B56.4
CodingCeleris-1Not measuredMarginNo overlapQwen3.5-122B-A10B72.0
ReasoningCeleris-1Not measuredMarginNo overlapQwen3.5-122B-A10B60.2
MultilingualCeleris-1Not measuredMarginNo overlapQwen3.5-122B-A10B82.2
MultimodalCeleris-1Not measuredMarginNo overlapQwen3.5-122B-A10B77.2
Inst. FollowingCeleris-1Not measuredMarginNo overlapQwen3.5-122B-A10B93.4

Decisive benchmark drivers

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

More
A · Celeris-1B · Qwen3.5-122B-A10B
  1. MMLU-Pro

    Knowledge
    Source ↗
    A 75.9%B 86.7%
    Winner: Qwen3.5-122B-A10BΔ 10.8
    MMLU-Pro: Celeris-1 scored 75.9%; Qwen3.5-122B-A10B scored 86.7%. Qwen3.5-122B-A10B wins this benchmark.

Operational comparison

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

MetricCeleris-1Qwen3.5-122B-A10BComparison
Input / output priceUSD per 1M tokensCeleris-1$2 input / $6 outputQwen3.5-122B-A10B$0 input / $0 outputQwen3.5-122B-A10B has the lower combined listed price.
Generation speedtokens per secondCeleris-1Not availableQwen3.5-122B-A10BNot availableA complete speed comparison is not available.
First-answer latencyseconds to first tokenCeleris-1Not availableQwen3.5-122B-A10BNot availableA complete latency comparison is not available.
Context windowmaximum listed tokensCeleris-18KQwen3.5-122B-A10B262KQwen3.5-122B-A10B lists the larger context window.

Benchmark Deep Dive

Agentic
BenchmarkCeleris-1Qwen3.5-122B-A10BResult
Terminal-Bench 2.0Source 49.4%Not comparable
BrowseCompSource 63.8%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 978Not comparable
Coding
BenchmarkCeleris-1Qwen3.5-122B-A10BResult
SWE-bench VerifiedSource 72%Not comparable
AA Coding IndexSource 45.7%Not comparable
AA-SciCodeSource 42.0%Not comparable
Reasoning
BenchmarkCeleris-1Qwen3.5-122B-A10BResult
LongBench v2Source 60.2%Not comparable
AA-LCRSource 66.7%Not comparable
CritPtSource 0.6%Not comparable
KnowledgeQwen3.5-122B-A10B wins
BenchmarkCeleris-1Qwen3.5-122B-A10BResult
MMLU-ProSource 75.9%86.7%Qwen3.5-122B-A10B leads
SuperGPQASource 67.1%Not comparable
GPQASource 86.6%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
Multilingual
BenchmarkCeleris-1Qwen3.5-122B-A10BResult
MMLU-ProXSource 82.2%Not comparable
Multimodal
BenchmarkCeleris-1Qwen3.5-122B-A10BResult
MMMUSource 83.9%Not comparable
MMVUSource 74.7%Not comparable
MathVisionSource 86.2%Not comparable
CharXivSource 77.2%Not comparable
V*Source 93.2%Not comparable
AA-MMMU-ProSource 75.0%Not comparable
Inst. Following
BenchmarkCeleris-1Qwen3.5-122B-A10BResult
IFEvalSource 93.4%Not comparable
AA-IFBenchSource 75.7%Not comparable
Frequently Asked Questions (2)

Which is better, Celeris-1 or Qwen3.5-122B-A10B?

Celeris-1 and Qwen3.5-122B-A10B are tied on the BenchAlign overall score, so the right pick depends on which category matters most for your use case.

Which is better for knowledge tasks, Celeris-1 or Qwen3.5-122B-A10B?

Qwen3.5-122B-A10B has the edge for knowledge tasks in this comparison, averaging 83.6 versus 75.9. Inside this category, MMLU-Pro is the benchmark that creates the most daylight between them.

Related Comparisons

Last updated: July 24, 2026

Choose a model with this week’s evidence

Join 2,000+ readers for ranking moves, pricing changes, and the claims that still need proof.

One email each week. Unsubscribe anytime.