Skip to main content

Model comparison

Claude 3 Haiku vs GLM-5-Turbo

Data verified

Head-to-head evidence from 11 shared benchmark results across 5 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.

21.39/100
Margin
45.5pts
winning →
66.89/100
0 category wins0 category wins

Public leaderboard positions: Claude 3 Haiku #194 (Supported); GLM-5-Turbo #24 (Supported). Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.

Evidence parity. Claude 3 Haiku and GLM-5-Turbo share 11 comparable benchmark results. 0 of 8 categories are comparable. 1 result is unique to Claude 3 Haiku; 2 to GLM-5-Turbo.

Updated July 23, 2026
Shared results
11
Claude 3 Haiku only
1
GLM-5-Turbo only
2
Comparable categories
0 / 8

Benchmark data for Claude 3 Haiku and GLM-5-Turbo is coming soon on BenchLM.

Confidence note. This is a partial-evidence comparison with 11 shared benchmark results across 5 evidence categories; 0 of 8 categories currently have scoreable aggregates for both models. Treat the verdict as directional until coverage is more balanced.

Why this result

BenchLM has partial data for these models, but not enough overlapping benchmark coverage to produce a fair score-level comparison yet.

GLM-5-Turbo is priced at $1.20 input / $4.00 output per 1M tokens, versus $0.25 input / $1.25 output per 1M tokens for Claude 3 Haiku.

Category breakdown

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

Operational comparison

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

MetricClaude 3 HaikuGLM-5-TurboComparison
Input / output priceUSD per 1M tokensClaude 3 Haiku$0.25 input / $1.25 outputGLM-5-Turbo$1.2 input / $4 outputClaude 3 Haiku has the lower combined listed price.
Generation speedtokens per secondClaude 3 Haiku138 tok/sGLM-5-TurboNot availableA complete speed comparison is not available.
First-answer latencyseconds to first tokenClaude 3 Haiku1.16 sGLM-5-TurboNot availableA complete latency comparison is not available.
Context windowmaximum listed tokensClaude 3 Haiku200KGLM-5-Turbo200KListed context windows are equal.

Benchmark Deep Dive

Agentic
BenchmarkClaude 3 HaikuGLM-5-TurboResult
τ²-bench resultsSource 21.1%98.5%GLM-5-Turbo leads
Claw-EvalSource 55.8%Not comparable
Coding
BenchmarkClaude 3 HaikuGLM-5-TurboResult
AA-SciCodeSource 18.6%43.6%GLM-5-Turbo leads
Reasoning
BenchmarkClaude 3 HaikuGLM-5-TurboResult
AA-LCRSource 21.0%60.7%GLM-5-Turbo leads
CritPtSource 0.0%0.3%GLM-5-Turbo leads
Knowledge
BenchmarkClaude 3 HaikuGLM-5-TurboResult
Artificial Analysis Intelligence IndexSource 3.9%38.1%GLM-5-Turbo leads
AA-GPQA DiamondSource 37.4%84.7%GLM-5-Turbo leads
AA-HLESource 3.9%25.4%GLM-5-Turbo leads
AA-Omniscience IndexSource -47.6%-15.1%GLM-5-Turbo leads
AA-Omniscience AccuracySource 17.2%29.0%GLM-5-Turbo leads
AA-Omniscience Hallucination RateSource 78.2%62.2%GLM-5-Turbo leads
Multimodal
BenchmarkClaude 3 HaikuGLM-5-TurboResult
AA-MMMU-ProSource 30.8%Not comparable
Design Arena WebsiteSource 1301Not comparable
Inst. Following
BenchmarkClaude 3 HaikuGLM-5-TurboResult
AA-IFBenchSource 36.1%73.2%GLM-5-Turbo leads
Frequently Asked Questions (3)

Can I compare Claude 3 Haiku and GLM-5-Turbo on BenchLM yet?

Not fully yet. BenchLM is tracking both models, but the sourced benchmark breakdown for this comparison is still coming soon.

Why does this comparison show “coming soon”?

BenchLM only shows category winners and benchmark-level calls when we have sourced results that can be compared fairly. For these models, the public benchmark coverage is not complete enough yet.

What data is available for Claude 3 Haiku and GLM-5-Turbo today?

Claude 3 Haiku: $0.25 input / $1.25 output per 1M tokens GLM-5-Turbo: $1.20 input / $4.00 output per 1M tokens Both model pages still include creator, context window, reasoning mode, and other metadata while benchmark coverage fills in.

Related Comparisons

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.