Skip to main content
BenchLM
Advertise

Model comparison

Claude Mythos 5 vs Qwen3.5-35B-A3B

Data verified

Head-to-head evidence from 5 shared benchmark results across 3 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.

83.93/100
Margin
27.0pts
← winning
56.97/100
2 category wins1 category wins

Public leaderboard positions: Claude Mythos 5 #1 (Supported); Qwen3.5-35B-A3B #72 (Supported). Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.

Evidence parity. Claude Mythos 5 and Qwen3.5-35B-A3B share 5 comparable benchmark results. 3 of 8 categories are comparable. 10 results are unique to Claude Mythos 5; 23 to Qwen3.5-35B-A3B.

Updated July 23, 2026
Shared results
5
Claude Mythos 5 only
10
Qwen3.5-35B-A3B only
23
Comparable categories
3 / 8

Pick Claude Mythos 5 if you want the stronger benchmark profile. Qwen3.5-35B-A3B 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 3 evidence categories; 3 of 8 categories currently have scoreable aggregates for both models. Treat the verdict as directional until coverage is more balanced.

Why this result

Claude Mythos 5 is clearly ahead on the BenchAlign aggregate, 83.93 to 56.97. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.

Claude Mythos 5's sharpest advantage is in agentic, where it averages 87 against 51. The single biggest benchmark swing on the page is Terminal-Bench 2.0, 88% to 40.5%. Qwen3.5-35B-A3B does hit back in knowledge, so the answer changes if that is the part of the workload you care about most.

Claude Mythos 5 is also the more expensive model on tokens at $10.00 input / $50.00 output per 1M tokens, versus $0.00 input / $0.00 output per 1M tokens for Qwen3.5-35B-A3B. That is roughly Infinityx on output cost alone. Claude Mythos 5 gives you the larger context window at 1M+, compared with 262K for Qwen3.5-35B-A3B.

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 Claude Mythos 5 and Qwen3.5-35B-A3B
CategoryClaude Mythos 5ΔQwen3.5-35B-A3B
AgenticClaude Mythos 587.0Margin← 36.0Qwen3.5-35B-A3B51.0
CodingClaude Mythos 589.7Margin← 29.1Qwen3.5-35B-A3B60.6
KnowledgeClaude Mythos 568.5Margin→ 13.1Qwen3.5-35B-A3B81.6
ReasoningClaude Mythos 5Not measuredMarginNo overlapQwen3.5-35B-A3B59.0
MathClaude Mythos 597.6MarginNo overlapQwen3.5-35B-A3BNot measured
MultilingualClaude Mythos 5Not measuredMarginNo overlapQwen3.5-35B-A3B81.0
MultimodalClaude Mythos 593.5MarginNo overlapQwen3.5-35B-A3BNot measured
Inst. FollowingClaude Mythos 5Not measuredMarginNo overlapQwen3.5-35B-A3B91.9

Decisive benchmark drivers

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

More
A · Claude Mythos 5B · Qwen3.5-35B-A3B

  1. Terminal-Bench 2.0

    Agentic
    Source ↗
    A 88%B 40.5%
    Winner: Claude Mythos 5Δ 47.5
    Terminal-Bench 2.0: Claude Mythos 5 scored 88%; Qwen3.5-35B-A3B scored 40.5%. Claude Mythos 5 wins this benchmark.

  2. OSWorld-Verified

    Agentic
    Source ↗
    A 85%B 54.5%
    Winner: Claude Mythos 5Δ 30.5
    OSWorld-Verified: Claude Mythos 5 scored 85%; Qwen3.5-35B-A3B scored 54.5%. Claude Mythos 5 wins this benchmark.

  3. BrowseComp

    Agentic
    Source ↗
    A 88%B 61%
    Winner: Claude Mythos 5Δ 27
    BrowseComp: Claude Mythos 5 scored 88%; Qwen3.5-35B-A3B scored 61%. Claude Mythos 5 wins this benchmark.

  4. SWE-bench Verified

    Coding
    Source ↗
    A 95.5%B 69.2%
    Winner: Claude Mythos 5Δ 26.3
    SWE-bench Verified: Claude Mythos 5 scored 95.5%; Qwen3.5-35B-A3B scored 69.2%. Claude Mythos 5 wins this benchmark.

  5. GPQA

    Knowledge
    Source ↗
    A 94.1%B 84.2%
    Winner: Claude Mythos 5Δ 9.9
    GPQA: Claude Mythos 5 scored 94.1%; Qwen3.5-35B-A3B scored 84.2%. Claude Mythos 5 wins this benchmark.

Operational comparison

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

MetricClaude Mythos 5Qwen3.5-35B-A3BComparison
Input / output priceUSD per 1M tokensClaude Mythos 5$10 input / $50 outputQwen3.5-35B-A3B$0 input / $0 outputQwen3.5-35B-A3B has the lower combined listed price.
Generation speedtokens per secondClaude Mythos 5Not availableQwen3.5-35B-A3BNot availableA complete speed comparison is not available.
First-answer latencyseconds to first tokenClaude Mythos 5Not availableQwen3.5-35B-A3BNot availableA complete latency comparison is not available.
Context windowmaximum listed tokensClaude Mythos 51M+Qwen3.5-35B-A3B262KClaude Mythos 5 lists the larger context window.

Benchmark Deep Dive

AgenticClaude Mythos 5 wins
BenchmarkClaude Mythos 5Qwen3.5-35B-A3BResult
Terminal-Bench 2.0Source 88%40.5%Claude Mythos 5 leads
OSWorld-VerifiedSource 85%54.5%Claude Mythos 5 leads
BrowseCompSource 88%61%Claude Mythos 5 leads
ExploitGymSource 17.5%—Not comparable
τ²-bench resultsSource —89.2%Not comparable
Gert LabsSource —28.96%Not comparable
CodingClaude Mythos 5 wins
BenchmarkClaude Mythos 5Qwen3.5-35B-A3BResult
SWE-bench VerifiedSource 95.5%69.2%Claude Mythos 5 leads
SWE-bench ProSource 80.3%—Not comparable
Terminal-Bench 2.0Source 88.0%—Not comparable
SWE-RebenchSource —53.7%Not comparable
AA-SciCodeSource —37.7%Not comparable
Reasoning
BenchmarkClaude Mythos 5Qwen3.5-35B-A3BResult
LongBench v2Source —59%Not comparable
AA-LCRSource —62.7%Not comparable
CritPtSource —0.9%Not comparable
KnowledgeQwen3.5-35B-A3B wins
BenchmarkClaude Mythos 5Qwen3.5-35B-A3BResult
GPQASource 94.1%84.2%Claude Mythos 5 leads
HLESource 64.5%—Not comparable
HLE w/o toolsSource 59%—Not comparable
MMLU-ProSource —85.3%Not comparable
SuperGPQASource —63.4%Not comparable
Artificial Analysis Intelligence IndexSource —29.3%Not comparable
AA-GPQA DiamondSource —84.5%Not comparable
AA-HLESource —19.7%Not comparable
AA-Omniscience IndexSource —-46.4%Not comparable
AA-Omniscience AccuracySource —20.5%Not comparable
AA-Omniscience Hallucination RateSource —84.0%Not comparable
Math
BenchmarkClaude Mythos 5Qwen3.5-35B-A3BResult
USAMO 2026Source 97.6%—Not comparable
Multilingual
BenchmarkClaude Mythos 5Qwen3.5-35B-A3BResult
SWE MultilingualSource 92.2%—Not comparable
MMLU-ProXSource —81%Not comparable
Multimodal
BenchmarkClaude Mythos 5Qwen3.5-35B-A3BResult
SWE-bench MultimodalSource 54.9%—Not comparable
CharXivSource 93.5%—Not comparable
CharXiv w/o toolsSource 88.9%—Not comparable
MMMUSource —81.4%Not comparable
MMVUSource —72.3%Not comparable
MathVisionSource —83.9%Not comparable
V*Source —92.7%Not comparable
AA-MMMU-ProSource —72.7%Not comparable
Inst. Following
BenchmarkClaude Mythos 5Qwen3.5-35B-A3BResult
IFEvalSource —91.9%Not comparable
AA-IFBenchSource —72.5%Not comparable
Frequently Asked Questions (4)

Which is better, Claude Mythos 5 or Qwen3.5-35B-A3B?

Claude Mythos 5 is ahead on BenchLM's BenchAlign leaderboard, 83.93 to 56.97. The biggest single separator in this matchup is Terminal-Bench 2.0, where the scores are 88% and 40.5%.

Which is better for knowledge tasks, Claude Mythos 5 or Qwen3.5-35B-A3B?

Qwen3.5-35B-A3B has the edge for knowledge tasks in this comparison, averaging 81.6 versus 68.5. Inside this category, GPQA is the benchmark that creates the most daylight between them.

Which is better for coding, Claude Mythos 5 or Qwen3.5-35B-A3B?

Claude Mythos 5 has the edge for coding in this comparison, averaging 89.7 versus 60.6. Inside this category, SWE-bench Verified is the benchmark that creates the most daylight between them.

Which is better for agentic tasks, Claude Mythos 5 or Qwen3.5-35B-A3B?

Claude Mythos 5 has the edge for agentic tasks in this comparison, averaging 87 versus 51. Inside this category, Terminal-Bench 2.0 is the benchmark that creates the most daylight between them.

Related Comparisons

Explore More

Anthropic Compare Pricing Methodology Find Your Best LLM Overall Rankings
Last updated: July 23, 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.