Z.ai: GLM 5.3

GLM-5.3 is a large-scale reasoning model from Z.ai, built for complex software engineering and long-horizon agent tasks. It supports text input and output with a 1M-token context window, and improves…

z-ai/glm-5.3-20260816

GLM-5.3 is a large-scale reasoning model from Z.ai, built for complex software engineering and long-horizon agent tasks. It supports text input and output with a 1M-token context window, and improves on GLM-5.2 in coding and in the balance between performance and token efficiency.

Reasoning is always on and cannot be disabled. Reasoning efforts low, high, and max are supported; max is the default.

Specifications & capabilities

Input
text
Output
text
Context window · tokens
1048576
Published model ID
GLM-5.3
Capabilities
promptCaching, reasoning, structuredOutputs, toolCalling
Description source ↗

Evaluations

81–97 / 97

OpenRouter Benchmarks

83.021%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · Cloudflare
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
13
sample size
2346
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
12
conditions
window
newest 90 days; successful visible nonzero runs
dataset
benchmark-leaderboard
benchmark
id
gpqa_diamond
name
GPQA Diamond
record_kind
aggregate
provider_name
Cloudflare
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-10-31T08:14:43.196Z
license_url
https://creativecommons.org/licenses/by/4.0/
methodology
Rows aggregate the benchmark-runs dataset exactly as the leaderboard pages do, over each benchmark's non-zero-score runs from the newest 90 days: accuracy is the task-weighted mean of per-run accuracy, accuracy_std_dev is the population standard deviation of per-run accuracy (null with a single run), and the per-task averages divide summed cost, generation time, and output tokens by summed tasks. Rows are ordered by accuracy descending, then model name.
metric_kind
GPQA Diamond · Accuracy
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
sample_size
2346
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/gpqa-diamond
measurements
accuracy
0.8302074978687126
accuracy_std_dev
0.018798024809619108
cost_per_task_usd
0.14303728772378516
output_tokens_per_task
32292.15643648764
latency_seconds_per_task
291.11502642796245
published_at
2026-11-16T01:46:02.653Z
snapshot_date
2026-11-16
source_revision
7ccaf17664252a3053dcb7df8ad0a46e72954a3b4524ddc1c8a7ef8a41b15568
transformations
TensorTrace preserves source results, converts latency to seconds and derives ranks within matching conditions. Individual runs have no derived rank.
metric_population
18
benchmark_evaluation
true
openrouter_permaslug
z-ai/glm-5.3-20260816
source model id
z-ai/glm-5.3-20260816
Original result ↗

OpenRouter Benchmarks

82.642%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · CoreWeave
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
14
sample size
771
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
4
conditions
window
newest 90 days; successful visible nonzero runs
dataset
benchmark-leaderboard
benchmark
id
gpqa_diamond
name
GPQA Diamond
record_kind
aggregate
provider_name
CoreWeave
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-10-31T08:38:15.143Z
license_url
https://creativecommons.org/licenses/by/4.0/
methodology
Rows aggregate the benchmark-runs dataset exactly as the leaderboard pages do, over each benchmark's non-zero-score runs from the newest 90 days: accuracy is the task-weighted mean of per-run accuracy, accuracy_std_dev is the population standard deviation of per-run accuracy (null with a single run), and the per-task averages divide summed cost, generation time, and output tokens by summed tasks. Rows are ordered by accuracy descending, then model name.
metric_kind
GPQA Diamond · Accuracy
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
sample_size
771
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/gpqa-diamond
measurements
accuracy
0.8264159143968872
accuracy_std_dev
0.0040531378816294926
cost_per_task_usd
0.13822659403372245
output_tokens_per_task
34382.090791180286
latency_seconds_per_task
272.7160363164721
published_at
2026-11-16T01:46:02.653Z
snapshot_date
2026-11-16
source_revision
7ccaf17664252a3053dcb7df8ad0a46e72954a3b4524ddc1c8a7ef8a41b15568
transformations
TensorTrace preserves source results, converts latency to seconds and derives ranks within matching conditions. Individual runs have no derived rank.
metric_population
27
benchmark_evaluation
true
openrouter_permaslug
z-ai/glm-5.3-20260816
source model id
z-ai/glm-5.3-20260816
Original result ↗

OpenRouter Benchmarks

85.565%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · GMICloud
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
15
sample size
2562
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
13
conditions
window
newest 90 days; successful visible nonzero runs
dataset
benchmark-leaderboard
benchmark
id
gpqa_diamond
name
GPQA Diamond
record_kind
aggregate
provider_name
GMICloud
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-11-10T06:48:32.349Z
license_url
https://creativecommons.org/licenses/by/4.0/
methodology
Rows aggregate the benchmark-runs dataset exactly as the leaderboard pages do, over each benchmark's non-zero-score runs from the newest 90 days: accuracy is the task-weighted mean of per-run accuracy, accuracy_std_dev is the population standard deviation of per-run accuracy (null with a single run), and the per-task averages divide summed cost, generation time, and output tokens by summed tasks. Rows are ordered by accuracy descending, then model name.
metric_kind
GPQA Diamond · Accuracy
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
sample_size
2562
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/gpqa-diamond
measurements
accuracy
0.8556465066354411
accuracy_std_dev
0.020339777161859043
cost_per_task_usd
0.08425942115534739
output_tokens_per_task
25603.65261514442
latency_seconds_per_task
311.6560573770492
published_at
2026-11-16T01:46:02.653Z
snapshot_date
2026-11-16
source_revision
7ccaf17664252a3053dcb7df8ad0a46e72954a3b4524ddc1c8a7ef8a41b15568
transformations
TensorTrace preserves source results, converts latency to seconds and derives ranks within matching conditions. Individual runs have no derived rank.
metric_population
26
benchmark_evaluation
true
openrouter_permaslug
z-ai/glm-5.3-20260816
source model id
z-ai/glm-5.3-20260816
Original result ↗

OpenRouter Benchmarks

84.465%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · SiliconFlow
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
15
sample size
1974
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
10
conditions
window
newest 90 days; successful visible nonzero runs
dataset
benchmark-leaderboard
benchmark
id
gpqa_diamond
name
GPQA Diamond
record_kind
aggregate
provider_name
SiliconFlow
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-10-31T08:32:53.731Z
license_url
https://creativecommons.org/licenses/by/4.0/
methodology
Rows aggregate the benchmark-runs dataset exactly as the leaderboard pages do, over each benchmark's non-zero-score runs from the newest 90 days: accuracy is the task-weighted mean of per-run accuracy, accuracy_std_dev is the population standard deviation of per-run accuracy (null with a single run), and the per-task averages divide summed cost, generation time, and output tokens by summed tasks. Rows are ordered by accuracy descending, then model name.
metric_kind
GPQA Diamond · Accuracy
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
sample_size
1974
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/gpqa-diamond
measurements
accuracy
0.8446468343465044
accuracy_std_dev
0.01783018725336332
cost_per_task_usd
0.0660811013171226
output_tokens_per_task
20336.286220871327
latency_seconds_per_task
245.35375430597773
published_at
2026-11-16T01:46:02.653Z
snapshot_date
2026-11-16
source_revision
7ccaf17664252a3053dcb7df8ad0a46e72954a3b4524ddc1c8a7ef8a41b15568
transformations
TensorTrace preserves source results, converts latency to seconds and derives ranks within matching conditions. Individual runs have no derived rank.
metric_population
34
benchmark_evaluation
true
openrouter_permaslug
z-ai/glm-5.3-20260816
source model id
z-ai/glm-5.3-20260816
Original result ↗

OpenRouter Benchmarks

83.838%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · Nebius
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
15
sample size
1386
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
7
conditions
window
newest 90 days; successful visible nonzero runs
dataset
benchmark-leaderboard
benchmark
id
gpqa_diamond
name
GPQA Diamond
record_kind
aggregate
provider_name
Nebius
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-10-31T08:45:52.318Z
license_url
https://creativecommons.org/licenses/by/4.0/
methodology
Rows aggregate the benchmark-runs dataset exactly as the leaderboard pages do, over each benchmark's non-zero-score runs from the newest 90 days: accuracy is the task-weighted mean of per-run accuracy, accuracy_std_dev is the population standard deviation of per-run accuracy (null with a single run), and the per-task averages divide summed cost, generation time, and output tokens by summed tasks. Rows are ordered by accuracy descending, then model name.
metric_kind
GPQA Diamond · Accuracy
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
sample_size
1386
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/gpqa-diamond
measurements
accuracy
0.8383837142857142
accuracy_std_dev
0.012139811903865117
cost_per_task_usd
0.1527254054834055
output_tokens_per_task
34470.08658008658
latency_seconds_per_task
220.5320505050505
published_at
2026-11-16T01:46:02.653Z
snapshot_date
2026-11-16
source_revision
7ccaf17664252a3053dcb7df8ad0a46e72954a3b4524ddc1c8a7ef8a41b15568
transformations
TensorTrace preserves source results, converts latency to seconds and derives ranks within matching conditions. Individual runs have no derived rank.
metric_population
25
benchmark_evaluation
true
openrouter_permaslug
z-ai/glm-5.3-20260816
source model id
z-ai/glm-5.3-20260816
Original result ↗

OpenRouter Benchmarks

83.413%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · Alibaba
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
16
sample size
3366
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
17
conditions
window
newest 90 days; successful visible nonzero runs
dataset
benchmark-leaderboard
benchmark
id
gpqa_diamond
name
GPQA Diamond
record_kind
aggregate
provider_name
Alibaba
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-10-31T08:41:02.403Z
license_url
https://creativecommons.org/licenses/by/4.0/
methodology
Rows aggregate the benchmark-runs dataset exactly as the leaderboard pages do, over each benchmark's non-zero-score runs from the newest 90 days: accuracy is the task-weighted mean of per-run accuracy, accuracy_std_dev is the population standard deviation of per-run accuracy (null with a single run), and the per-task averages divide summed cost, generation time, and output tokens by summed tasks. Rows are ordered by accuracy descending, then model name.
metric_kind
GPQA Diamond · Accuracy
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
sample_size
3366
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/gpqa-diamond
measurements
accuracy
0.8341254705882353
accuracy_std_dev
0.010649185432078013
cost_per_task_usd
0.25025737195484254
output_tokens_per_task
40253.002970885325
latency_seconds_per_task
384.70965567439094
published_at
2026-11-16T01:46:02.653Z
snapshot_date
2026-11-16
source_revision
7ccaf17664252a3053dcb7df8ad0a46e72954a3b4524ddc1c8a7ef8a41b15568
transformations
TensorTrace preserves source results, converts latency to seconds and derives ranks within matching conditions. Individual runs have no derived rank.
metric_population
29
benchmark_evaluation
true
openrouter_permaslug
z-ai/glm-5.3-20260816
source model id
z-ai/glm-5.3-20260816
Original result ↗

OpenRouter Benchmarks

86.994%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · DeepInfra
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
17
sample size
1894
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
10
conditions
window
newest 90 days; successful visible nonzero runs
dataset
benchmark-leaderboard
benchmark
id
gpqa_diamond
name
GPQA Diamond
record_kind
aggregate
provider_name
DeepInfra
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-10-31T08:45:16.438Z
license_url
https://creativecommons.org/licenses/by/4.0/
methodology
Rows aggregate the benchmark-runs dataset exactly as the leaderboard pages do, over each benchmark's non-zero-score runs from the newest 90 days: accuracy is the task-weighted mean of per-run accuracy, accuracy_std_dev is the population standard deviation of per-run accuracy (null with a single run), and the per-task averages divide summed cost, generation time, and output tokens by summed tasks. Rows are ordered by accuracy descending, then model name.
metric_kind
GPQA Diamond · Accuracy
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
sample_size
1894
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/gpqa-diamond
measurements
accuracy
0.8699400591341078
accuracy_std_dev
0.01681928135474285
cost_per_task_usd
0.07171590390707498
output_tokens_per_task
23628.092397043296
latency_seconds_per_task
408.5245543822598
published_at
2026-11-16T01:46:02.653Z
snapshot_date
2026-11-16
source_revision
7ccaf17664252a3053dcb7df8ad0a46e72954a3b4524ddc1c8a7ef8a41b15568
transformations
TensorTrace preserves source results, converts latency to seconds and derives ranks within matching conditions. Individual runs have no derived rank.
metric_population
71
benchmark_evaluation
true
openrouter_permaslug
z-ai/glm-5.3-20260816
source model id
z-ai/glm-5.3-20260816
Original result ↗

OpenRouter Benchmarks

73.333%

source id
openRouterBenchmarks
leaderboard name
tau2-bench Verified Airline · Accuracy · Amazon Bedrock
dimension
agentic
metric
tau2-bench Verified Airline · Accuracy
rank
17
sample size
150
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
3
conditions
window
newest 90 days; successful visible nonzero runs
dataset
benchmark-leaderboard
benchmark
id
tau_bench_verified_airline
name
tau2-bench Verified Airline
record_kind
aggregate
provider_name
Amazon Bedrock
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-10-31T12:28:59.426Z
license_url
https://creativecommons.org/licenses/by/4.0/
methodology
Rows aggregate the benchmark-runs dataset exactly as the leaderboard pages do, over each benchmark's non-zero-score runs from the newest 90 days: accuracy is the task-weighted mean of per-run accuracy, accuracy_std_dev is the population standard deviation of per-run accuracy (null with a single run), and the per-task averages divide summed cost, generation time, and output tokens by summed tasks. Rows are ordered by accuracy descending, then model name.
metric_kind
tau2-bench Verified Airline · Accuracy
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
sample_size
150
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/tau2-bench-airline
measurements
accuracy
0.7333333333333333
accuracy_std_dev
0.024944382578492966
cost_per_task_usd
0.09260214
output_tokens_per_task
4729.153333333334
latency_seconds_per_task
326.5589733333333
published_at
2026-11-16T01:46:02.653Z
snapshot_date
2026-11-16
source_revision
7ccaf17664252a3053dcb7df8ad0a46e72954a3b4524ddc1c8a7ef8a41b15568
transformations
TensorTrace preserves source results, converts latency to seconds and derives ranks within matching conditions. Individual runs have no derived rank.
metric_population
31
benchmark_evaluation
true
openrouter_permaslug
z-ai/glm-5.3-20260816
source model id
z-ai/glm-5.3-20260816
Original result ↗

OpenRouter Benchmarks

83.178%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · Parasail
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
19
sample size
2149
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
11
conditions
window
newest 90 days; successful visible nonzero runs
dataset
benchmark-leaderboard
benchmark
id
gpqa_diamond
name
GPQA Diamond
record_kind
aggregate
provider_name
Parasail
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-11-05T00:00:09.943Z
license_url
https://creativecommons.org/licenses/by/4.0/
methodology
Rows aggregate the benchmark-runs dataset exactly as the leaderboard pages do, over each benchmark's non-zero-score runs from the newest 90 days: accuracy is the task-weighted mean of per-run accuracy, accuracy_std_dev is the population standard deviation of per-run accuracy (null with a single run), and the per-task averages divide summed cost, generation time, and output tokens by summed tasks. Rows are ordered by accuracy descending, then model name.
metric_kind
GPQA Diamond · Accuracy
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
sample_size
2149
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/gpqa-diamond
measurements
accuracy
0.8317821610051186
accuracy_std_dev
0.013001907940092663
cost_per_task_usd
0.15365638017682642
output_tokens_per_task
34707.94276407632
latency_seconds_per_task
272.02546998604004
published_at
2026-11-16T01:46:02.653Z
snapshot_date
2026-11-16
source_revision
7ccaf17664252a3053dcb7df8ad0a46e72954a3b4524ddc1c8a7ef8a41b15568
transformations
TensorTrace preserves source results, converts latency to seconds and derives ranks within matching conditions. Individual runs have no derived rank.
metric_population
36
benchmark_evaluation
true
openrouter_permaslug
z-ai/glm-5.3-20260816
source model id
z-ai/glm-5.3-20260816
Original result ↗

OpenRouter Benchmarks

84.084%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · Together
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
20
sample size
2177
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
11
conditions
window
newest 90 days; successful visible nonzero runs
dataset
benchmark-leaderboard
benchmark
id
gpqa_diamond
name
GPQA Diamond
record_kind
aggregate
provider_name
Together
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-10-31T09:21:48.528Z
license_url
https://creativecommons.org/licenses/by/4.0/
methodology
Rows aggregate the benchmark-runs dataset exactly as the leaderboard pages do, over each benchmark's non-zero-score runs from the newest 90 days: accuracy is the task-weighted mean of per-run accuracy, accuracy_std_dev is the population standard deviation of per-run accuracy (null with a single run), and the per-task averages divide summed cost, generation time, and output tokens by summed tasks. Rows are ordered by accuracy descending, then model name.
metric_kind
GPQA Diamond · Accuracy
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
sample_size
2177
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/gpqa-diamond
measurements
accuracy
0.8408359820854386
accuracy_std_dev
0.011558364030339736
cost_per_task_usd
0.14478948920532844
output_tokens_per_task
32673.29122645843
latency_seconds_per_task
203.0692958199357
published_at
2026-11-16T01:46:02.653Z
snapshot_date
2026-11-16
source_revision
7ccaf17664252a3053dcb7df8ad0a46e72954a3b4524ddc1c8a7ef8a41b15568
transformations
TensorTrace preserves source results, converts latency to seconds and derives ranks within matching conditions. Individual runs have no derived rank.
metric_population
25
benchmark_evaluation
true
openrouter_permaslug
z-ai/glm-5.3-20260816
source model id
z-ai/glm-5.3-20260816
Original result ↗

OpenRouter Benchmarks

81.986%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · StreamLake
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
20
sample size
198
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
1
conditions
window
newest 90 days; successful visible nonzero runs
dataset
benchmark-leaderboard
benchmark
id
gpqa_diamond
name
GPQA Diamond
record_kind
aggregate
provider_name
StreamLake
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-09-07T06:00:57.184Z
license_url
https://creativecommons.org/licenses/by/4.0/
methodology
Rows aggregate the benchmark-runs dataset exactly as the leaderboard pages do, over each benchmark's non-zero-score runs from the newest 90 days: accuracy is the task-weighted mean of per-run accuracy, accuracy_std_dev is the population standard deviation of per-run accuracy (null with a single run), and the per-task averages divide summed cost, generation time, and output tokens by summed tasks. Rows are ordered by accuracy descending, then model name.
metric_kind
GPQA Diamond · Accuracy
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
sample_size
198
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/gpqa-diamond
measurements
accuracy
0.819865
cost_per_task_usd
0.13854771717171718
output_tokens_per_task
31240.651515151516
latency_seconds_per_task
540.8241212121212
published_at
2026-11-16T01:46:02.653Z
snapshot_date
2026-11-16
source_revision
7ccaf17664252a3053dcb7df8ad0a46e72954a3b4524ddc1c8a7ef8a41b15568
transformations
TensorTrace preserves source results, converts latency to seconds and derives ranks within matching conditions. Individual runs have no derived rank.
metric_population
30
benchmark_evaluation
true
openrouter_permaslug
z-ai/glm-5.3-20260816
source model id
z-ai/glm-5.3-20260816
Original result ↗

OpenRouter Benchmarks

83.109%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · Amazon Bedrock
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
23
sample size
594
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
3
conditions
window
newest 90 days; successful visible nonzero runs
dataset
benchmark-leaderboard
benchmark
id
gpqa_diamond
name
GPQA Diamond
record_kind
aggregate
provider_name
Amazon Bedrock
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-10-31T08:09:31.329Z
license_url
https://creativecommons.org/licenses/by/4.0/
methodology
Rows aggregate the benchmark-runs dataset exactly as the leaderboard pages do, over each benchmark's non-zero-score runs from the newest 90 days: accuracy is the task-weighted mean of per-run accuracy, accuracy_std_dev is the population standard deviation of per-run accuracy (null with a single run), and the per-task averages divide summed cost, generation time, and output tokens by summed tasks. Rows are ordered by accuracy descending, then model name.
metric_kind
GPQA Diamond · Accuracy
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
sample_size
594
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/gpqa-diamond
measurements
accuracy
0.8310883333333333
accuracy_std_dev
0.009654676391377534
cost_per_task_usd
0.20155932323232326
output_tokens_per_task
37910.23063973064
latency_seconds_per_task
266.665861952862
published_at
2026-11-16T01:46:02.653Z
snapshot_date
2026-11-16
source_revision
7ccaf17664252a3053dcb7df8ad0a46e72954a3b4524ddc1c8a7ef8a41b15568
transformations
TensorTrace preserves source results, converts latency to seconds and derives ranks within matching conditions. Individual runs have no derived rank.
metric_population
32
benchmark_evaluation
true
openrouter_permaslug
z-ai/glm-5.3-20260816
source model id
z-ai/glm-5.3-20260816
Original result ↗

OpenRouter Benchmarks

83.869%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · Novita
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
26
sample size
2178
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
11
conditions
window
newest 90 days; successful visible nonzero runs
dataset
benchmark-leaderboard
benchmark
id
gpqa_diamond
name
GPQA Diamond
record_kind
aggregate
provider_name
Novita
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-10-31T08:36:48.545Z
license_url
https://creativecommons.org/licenses/by/4.0/
methodology
Rows aggregate the benchmark-runs dataset exactly as the leaderboard pages do, over each benchmark's non-zero-score runs from the newest 90 days: accuracy is the task-weighted mean of per-run accuracy, accuracy_std_dev is the population standard deviation of per-run accuracy (null with a single run), and the per-task averages divide summed cost, generation time, and output tokens by summed tasks. Rows are ordered by accuracy descending, then model name.
metric_kind
GPQA Diamond · Accuracy
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
sample_size
2178
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/gpqa-diamond
measurements
accuracy
0.8386898181818183
accuracy_std_dev
0.022320469186247312
cost_per_task_usd
0.1030642401285583
output_tokens_per_task
35835.99357208448
latency_seconds_per_task
384.60890220385676
published_at
2026-11-16T01:46:02.653Z
snapshot_date
2026-11-16
source_revision
7ccaf17664252a3053dcb7df8ad0a46e72954a3b4524ddc1c8a7ef8a41b15568
transformations
TensorTrace preserves source results, converts latency to seconds and derives ranks within matching conditions. Individual runs have no derived rank.
metric_population
59
benchmark_evaluation
true
openrouter_permaslug
z-ai/glm-5.3-20260816
source model id
z-ai/glm-5.3-20260816
Original result ↗

OpenRouter Benchmarks

85.37%

source id
openRouterBenchmarks
leaderboard name
GPQA Diamond · Accuracy · auto-routing
dimension
knowledge
metric
GPQA Diamond · Accuracy
rank
57
sample size
1980
aggregation scope
status
published
harness
url
https://github.com/OpenRouterTeam/benchmark-harness
name
OpenRouter benchmark harness
direction
higherIsBetter
run_count
10
conditions
window
newest 90 days; successful visible nonzero runs
dataset
benchmark-leaderboard
benchmark
id
gpqa_diamond
name
GPQA Diamond
record_kind
aggregate
provider_name
auto-routing
attribution
Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter.
last_run_at
2026-10-31T08:16:11.390Z
license_url
https://creativecommons.org/licenses/by/4.0/
methodology
Rows aggregate the benchmark-runs dataset exactly as the leaderboard pages do, over each benchmark's non-zero-score runs from the newest 90 days: accuracy is the task-weighted mean of per-run accuracy, accuracy_std_dev is the population standard deviation of per-run accuracy (null with a single run), and the per-task averages divide summed cost, generation time, and output tokens by summed tasks. Rows are ordered by accuracy descending, then model name.
metric_kind
GPQA Diamond · Accuracy
rank_method
competition rank among reported evaluations in identical conditions, including unlinked observations
sample_size
1980
source_file
benchmark-leaderboard.json
source_page
https://openrouter.ai/benchmarks/gpqa-diamond
measurements
accuracy
0.8537037
accuracy_std_dev
0.012630753865466617
cost_per_task_usd
0.10325877878787879
output_tokens_per_task
31795.09494949495
latency_seconds_per_task
455.2946515151515
published_at
2026-11-16T01:46:02.653Z
snapshot_date
2026-11-16
source_revision
7ccaf17664252a3053dcb7df8ad0a46e72954a3b4524ddc1c8a7ef8a41b15568
transformations
TensorTrace preserves source results, converts latency to seconds and derives ranks within matching conditions. Individual runs have no derived rank.
metric_population
132
benchmark_evaluation
true
openrouter_permaslug
z-ai/glm-5.3-20260816
source model id
z-ai/glm-5.3-20260816
Original result ↗

Published prices

1–20 / 124

@cf/zai-org/glm-5.3 ↗

$0.26

Inference provider
cloudflareWorkersAI
Platform
Cloudflare Workers AI
Unit
input_tokens_per_million
Currency
USD
Billing conditions
Billing label
per M cached input tokens

@cf/zai-org/glm-5.3 ↗

$4.40

Inference provider
cloudflareWorkersAI
Platform
Cloudflare Workers AI
Unit
output_tokens_per_million
Currency
USD
Billing conditions
Billing label
per M output tokens

z-ai/glm-5.3 ↗

$0.07

Inference provider
openRouter
Platform
OpenRouter
Unit
cache_read_tokens_per_million
Currency
USD
Billing conditions

z-ai/glm-5.3 ↗

$0.07

Inference provider
openRouter
Platform
OpenRouter
Unit
input_tokens_per_million
Currency
USD
Billing conditions

z-ai/glm-5.3 ↗

$1.40

Inference provider
openRouter
Platform
OpenRouter
Unit
output_tokens_per_million
Currency
USD
Billing conditions

z-ai/glm-5.3:batch ↗

$0.10

Inference provider
openRouter
Platform
OpenRouter
Unit
cache_read_tokens_per_million
Currency
USD
Billing conditions
mode
batch

z-ai/glm-5.3:batch ↗

$0.45

Inference provider
openRouter
Platform
OpenRouter
Unit
input_tokens_per_million
Currency
USD
Billing conditions
mode
batch

z-ai/glm-5.3:batch ↗

$2.00

Inference provider
openRouter
Platform
OpenRouter
Unit
output_tokens_per_million
Currency
USD
Billing conditions
mode
batch

~z-ai/glm-latest ↗

$0.07

Inference provider
openRouter
Platform
OpenRouter
Unit
cache_read_tokens_per_million
Currency
USD
Billing conditions
alias_target
name
Z.ai: GLM 5.3
slug
z-ai/glm-5.3
routing_alias
true

~z-ai/glm-latest ↗

$0.07

Inference provider
openRouter
Platform
OpenRouter
Unit
input_tokens_per_million
Currency
USD
Billing conditions
alias_target
name
Z.ai: GLM 5.3
slug
z-ai/glm-5.3
routing_alias
true

~z-ai/glm-latest ↗

$1.40

Inference provider
openRouter
Platform
OpenRouter
Unit
output_tokens_per_million
Currency
USD
Billing conditions
alias_target
name
Z.ai: GLM 5.3
slug
z-ai/glm-5.3
routing_alias
true

zai-org/GLM-5.3 ↗

$0.26

Inference provider
siliconFlow
Platform
SiliconFlow
Unit
cache_read_tokens_per_million
Currency
USD
Billing conditions
Billing label
Cached input tokens (per 1M)

zai-org/GLM-5.3 ↗

$1.40

Inference provider
siliconFlow
Platform
SiliconFlow
Unit
input_tokens_per_million
Currency
USD
Billing conditions
Billing label
Input tokens (per 1M)

zai-org/GLM-5.3 ↗

$4.40

Inference provider
siliconFlow
Platform
SiliconFlow
Unit
output_tokens_per_million
Currency
USD
Billing conditions
Billing label
Output tokens (per 1M)

glm-5.3 ↗

$0.12

Inference provider
supernova
Platform
Supernova Labs
Unit
cache_read_tokens_per_million
Currency
USD
Billing conditions
lane
standard
billing_scope
Public standard rates; account terms may differ
standard_multiplier
0.45
Billing label
Cached input tokens (per 1M)

glm-5.3 ↗

$0.63

Inference provider
supernova
Platform
Supernova Labs
Unit
input_tokens_per_million
Currency
USD
Billing conditions
lane
standard
billing_scope
Public standard rates; account terms may differ
standard_multiplier
0.45
Billing label
Input tokens (per 1M)

glm-5.3 ↗

$1.98

Inference provider
supernova
Platform
Supernova Labs
Unit
output_tokens_per_million
Currency
USD
Billing conditions
lane
standard
billing_scope
Public standard rates; account terms may differ
standard_multiplier
0.45
Billing label
Output tokens (per 1M)

glm-5.3-fast ↗

$0.52

Inference provider
supernova
Platform
Supernova Labs
Unit
cache_read_tokens_per_million
Currency
USD
Billing conditions
lane
priority
billing_scope
Public standard rates; account terms may differ
standard_multiplier
1
Billing label
Cached input tokens (per 1M)

glm-5.3-fast ↗

$2.80

Inference provider
supernova
Platform
Supernova Labs
Unit
input_tokens_per_million
Currency
USD
Billing conditions
lane
priority
billing_scope
Public standard rates; account terms may differ
standard_multiplier
1
Billing label
Input tokens (per 1M)

glm-5.3-fast ↗

$8.80

Inference provider
supernova
Platform
Supernova Labs
Unit
output_tokens_per_million
Currency
USD
Billing conditions
lane
priority
billing_scope
Public standard rates; account terms may differ
standard_multiplier
1
Billing label
Output tokens (per 1M)

Provider specifications & access

1–5 / 5

glm-5.3 ↗

source model id
@cf/zai-org/glm-5.3
hosting
cloudflare-hosted
platforms
  • Cloudflare Workers AI
context length
1048576
architecture
input_modalities
  • text
output_modalities
  • text
capabilities
  • toolCalling

GLM-5.3 ↗

source model id
glm-5.3
hosting
supernova-hosted
platforms
  • Supernova Labs
architecture
input_modalities
  • text
output_modalities
  • text
capabilities
  • reasoning
  • promptCaching

GLM-5.3 Fast ↗

source model id
glm-5.3-fast
hosting
supernova-hosted
platforms
  • Supernova Labs
architecture
input_modalities
  • text
output_modalities
  • text
capabilities
  • reasoning
  • promptCaching

GLM 5.3 ↗

source model id
zai/glm-5.3
hosting
vercel-gateway
platforms
  • Vercel AI Gateway
context length
1000000
max output tokens
1000000
architecture
input_modalities
  • text
output_modalities
  • text
capabilities
  • promptCaching
  • reasoning
  • structuredOutputs
  • toolCalling
regions
  • us

GLM-5.3 ↗

source model id
zai-org/GLM-5.3
hosting
siliconflow-hosted
platforms
  • SiliconFlow
max output tokens
262144
architecture
input_modalities
  • text
output_modalities
  • text
capabilities
  • promptCaching
  • reasoning
  • structuredOutputs
  • toolCalling

Source specification observations

1–10 / 10

openRouter ↗

field
context_length
value
1048576
normalized value
1048576
scope
source-model
source id
openRouter
confidence
1

openRouter ↗

field
context_length
value
1048576
normalized value
1048576
scope
source-model
source id
openRouter
confidence
1

openRouter ↗

field
context_length
value
1048576
normalized value
1048576
scope
source-model
source id
openRouter
confidence
1

siliconFlow ↗

field
context_length
value
1049K
scope
source-model
source id
siliconFlow
confidence
1

vercelAIGatewayCatalog ↗

field
context_length
value
1000000
normalized value
1000000
scope
source-model
source id
vercelAIGatewayCatalog
confidence
1

openRouter ↗

field
max_output_tokens
value
943718
normalized value
943718
scope
source-model
source id
openRouter
confidence
1

openRouter ↗

field
max_output_tokens
value
131072
normalized value
131072
scope
source-model
source id
openRouter
confidence
1

openRouter ↗

field
max_output_tokens
value
943718
normalized value
943718
scope
source-model
source id
openRouter
confidence
1

siliconFlow ↗

field
max_output_tokens
value
262K
scope
source-model
source id
siliconFlow
confidence
1

vercelAIGatewayCatalog ↗

field
max_output_tokens
value
1000000
normalized value
1000000
scope
source-model
source id
vercelAIGatewayCatalog
confidence
1

Original sources

1–20 / 23

SiliconFlow · Model page ↗

z.ai · Model reference ↗

Hugging Face · Model repository ↗

Cloudflare AI · Model page ↗

OpenRouter · Model page ↗

OpenRouter · Model page ↗

SiliconFlow · Playground ↗

Vercel AI Gateway · Model page ↗

SiliconFlow · API documentation ↗

Supernova · Model documentation ↗

Supernova · Pricing ↗

arXiv · Paper · 2602.15763 ↗

GitHub · Source snapshot ↗

Cloudflare AI · Catalog ↗

Epoch AI · Catalog ↗

Hugging Face · Catalog ↗

Hugging Face · Catalog ↗

OpenRouter · Catalog ↗

OpenRouter · Catalog ↗

SiliconFlow · Catalog ↗

Source links

1–11 / 11

Hugging Face model card

1–1 / 1

zai-org/GLM-5.3 ↗

downloads
1761619
likes
2455
pipeline tag
text-generation
library name
transformers
License
other

Explore more models

Models

Loading
EVIDENCE / COMPARISON

Benchmarks

Explore every benchmark

Loading

These charts contain model-linked results. The charts above also include unlinked source profiles. Compare scores within each evaluation. Model quality, usage, transparency and system performance measure different things.

Install TensorTrace