Z.ai (GLM)
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...
Cost / 1M tokens
Low usage$0.07
input
$0.069
cached input
$1.40
output
Context Window
1M
Up to 944K max output
Architecture
753B total MoE
text → text
Quickstart & API
z-ai/glm-5.3Best for & Strengths
Key Strengths
Recommended Workloads in ChatOSS
- Demanding software engineering
- Agentic code review
- Self-hosted enterprise AI
Verified Evaluations
Benchmarks & Frontier Comparison
How GLM 5.3 compares against leading open-weight and proprietary frontier models across independent agentic coding and reasoning suites.
| Benchmark (Metric) | GLM 5.3 | DeepSeek V4.1 Flash | Kimi K3.1 Pro | Claude Opus 5.5 | Gemini 4 Argon |
|---|---|---|---|---|---|
| Agentic & Software Engineering | |||||
| ChatOSS Agentic Coding Index (%) | 58.3% | 63.2% | 65% | 74.1% | 73.9% |
| SWE-Bench Verified (%) | 95.4% | 91.5% | 94.8% | 97% | 94% |
| DeepSWE v1.1 (Pass@1 %) | 69% | 77% | 73.5% | 74.2% | 77.9% |
| Vibe Code Bench v1.1 (%) | 78.13% | 87.7% | 88.4% | 90.3% | 91.9% |
| Terminal-Bench 4.0 (%) | 42% | 42.3% | 46.2% | 65.2% | 57.6% |
| Code Migration (%) | 44.22% | 45.6% | 52% | 66.7% | 68.2% |
| Intelligence & Token Economics | |||||
| Artificial Analysis Intelligence Index (0–100) | 31.7 | 39.5 | 55.1 | 57.6 | 52.6 |
| Input Price (/ 1M tokens) | $0.07 | $0.15 | $2.00 | $4.00 | $4.00 |
| Output Price (/ 1M tokens) | $1.40 | $0.60 | $10.00 | $20.00 | $20.00 |
ChatOSS Agentic Coding Index
ChatOSS Benchmark Audit · 4-benchmark composite mean
SWE-Bench Verified
Vals AI · Terminus 2
DeepSWE v1.1
DeepSWE / Official Audit · Mini-SWE-Agent
Vibe Code Bench v1.1
Vals.ai · OpenHands
Terminal-Bench 4.0
Vals.ai / Artificial Analysis · Terminus 2
Terminal-Bench 2.1
Vals.ai · Terminus 2
Code Migration
Vals.ai · Vals evaluation harness
AutomationBench
Artificial Analysis v4.3
Artificial Analysis Intelligence Index
Artificial Analysis
GPQA Diamond (Pass@1)
Benchmark Audit
Humanity's Last Exam (Pass@1)
Artificial Analysis / Audit
Design Arena: 3D
Design Arena · Rank #18 (56.7% win)
Design Arena: ASCIIART
Design Arena · Rank #23 (47.2% win)
Design Arena: Code Generation
Design Arena · Rank #19 (52.7% win)
Design Arena: Data Visualization
Design Arena · Rank #34 (50.7% win)
Active Provider Endpoints (47)
Upstream inference providers serving GLM 5.3, with per-provider pricing, context limits, quantization, and 24-hour uptime.
| Provider | Input / 1M | Output / 1M | Cache Read / 1M | Context | Max Output | Quantization | Uptime (24h) |
|---|---|---|---|---|---|---|---|
Wafercache wafer/us | $0.07 | $1.40 | $0.069 | 1M | 944K | fp4 | 100.0% |
Relacecache relace/fp4 | $0.07 | $4.38 | $0.07 | 1M | 131K | fp4 | 100.0% |
InferenceNetcache inference-net/us | $0.08 | $3.60 | $0.076 | 1M | 944K | fp4 | 100.0% |
InferenceNetcache inference-net/fp4 | $0.08 | $3.60 | $0.076 | 1M | 944K | fp4 | 100.0% |
Rekacache reka | $0.15 | $3.00 | $0.12 | 1M | 262K | — | 97.9% |
Wafercache wafer/fp4 | $0.179 | $1.40 | $0.178 | 1M | 944K | fp4 | 100.0% |
AkashMLcache akashml/fp4 | $0.19 | $4.40 | $0.19 | 1M | 131K | fp4 | 100.0% |
PrimeIntellectcache primeintellect | $0.40 | $2.46 | $0.074 | 1M | 131K | — | 100.0% |
Makoracache makora/fp4 | $0.40 | $3.20 | $0.16 | 1M | 128K | fp4 | 99.8% |
DeepInfracache deepinfra/fp4 | $0.563 | $2.50 | $0.125 | 1M | 131K | fp4 | 99.4% |
Tenstorrentcache tenstorrent/fp8 | $0.60 | $2.50 | $0.10 | 1M | 131K | fp8 | 98.7% |
Decartcache decart/fp4 | $0.61 | $1.89 | $0.11 | 1M | 944K | fp4 | 100.0% |
Novitacache novita/fp8 | $0.63 | $1.98 | $0.117 | 1M | 131K | fp8 | 100.0% |
Morphcache morph/fp8 | $0.699 | $2.15 | $0.236 | 1M | 944K | fp8 | 99.8% |
SiliconFlowcache siliconflow/fp8 | $0.70 | $2.20 | $0.13 | 1M | 262K | fp8 | 100.0% |
GMICloudcache gmicloud/fp8 | $0.70 | $2.20 | $0.13 | 1M | 944K | fp8 | 100.0% |
Cloudflarecache cloudflare | $0.70 | $2.20 | $0.13 | 1M | 944K | — | 100.0% |
CoreWeavecache coreweave/nvfp4 | $0.75 | $3.00 | $0.19 | 1M | 944K | nvfp4 | 99.6% |
Sail Researchcache sail-research/us | $0.75 | $3.00 | $0.12 | 1M | 944K | fp8 | 100.0% |
DigitalOceancache digitalocean | $0.91 | $2.86 | $0.169 | 1M | 128K | — | 99.9% |
Fireworkscache fireworks/spot | $0.98 | $3.08 | $0.182 | 1M | 944K | — | 93.8% |
Inceptroncache inceptron/fp4 | $1.00 | $4.40 | $0.14 | 1M | 944K | fp4 | 100.0% |
Sail Researchcache sail-research/fp8 | $1.05 | $4.20 | $0.168 | 1M | 944K | fp8 | 100.0% |
Rivercache river | $1.12 | $3.52 | $0.208 | 1M | 131K | — | 100.0% |
Alibabacache alibaba | $1.19 | $3.74 | $0.238 | 1M | 131K | — | 100.0% |
Friendlicache friendli | $1.26 | $3.96 | $0.234 | 1M | 944K | — | 100.0% |
SambaNovacache sambanova | $1.40 | $4.40 | $0.07 | 1M | 944K | — | 99.5% |
Mistralcache mistral/zdr | $1.40 | $4.40 | $0.14 | 1M | 131K | nvfp4 | 100.0% |
Baiducache baidu/fp8 | $1.40 | $4.40 | $0.26 | 1M | 131K | fp8 | 99.4% |
BaseTencache baseten/fp4/us | $1.40 | $4.40 | $0.14 | 1M | 262K | fp4 | 98.5% |
Nebiuscache nebius/fp4 | $1.40 | $4.40 | $0.26 | 1M | 944K | fp4 | 96.9% |
Crusoecache crusoe/fp4 | $1.40 | $4.40 | $0.26 | 1M | 944K | fp4 | 99.1% |
Venicecache venice | $1.40 | $4.40 | $0.26 | 1M | 131K | — | 97.6% |
Togethercache together | $1.40 | $4.40 | $0.26 | 1M | 944K | — | 99.8% |
Parasailcache parasail/fp8 | $1.40 | $4.40 | $0.26 | 1M | 944K | fp8 | 99.6% |
Modalcache modal | $1.40 | $4.40 | $0.26 | 1M | 944K | — | 100.0% |
BaseTencache baseten/fp4 | $1.40 | $4.40 | $0.14 | 1M | 262K | fp4 | 98.5% |
Fireworkscache fireworks | $1.40 | $4.40 | $0.26 | 1M | 944K | — | 98.6% |
AtlasCloudcache atlas-cloud/fp8 | $1.40 | $4.40 | $0.26 | 1M | 131K | fp8 | 100.0% |
Z.AIcache z-ai/fp8 | $1.40 | $4.40 | $0.26 | 1M | 131K | fp8 | 100.0% |
Mistralcache mistral/nvfp4/eu | $1.54 | $4.84 | $0.154 | 1M | 131K | nvfp4 | 100.0% |
Amazon Bedrock amazon-bedrock/us-east-1 | $1.68 | $5.28 | $0.312 | 1M | 131K | — | 99.9% |
Parasailcache parasail/fast | $2.10 | $6.60 | $0.26 | 1M | 944K | fp8 | 96.3% |
Fireworkscache fireworks/fast | $2.10 | $6.60 | $0.39 | 1M | 944K | — | 99.4% |
BaseTencache baseten/us/fast | $2.10 | $6.60 | $0.21 | 1M | 262K | fp8 | 97.9% |
BaseTencache baseten/fast | $2.10 | $6.60 | $0.21 | 1M | 262K | fp8 | 98.9% |
Alibabacache alibaba/prime | $2.80 | $8.80 | $0.56 | 1M | 131K | — | 100.0% |
Supported API Parameters
Parameters accepted by this model when called through ChatOSS or OpenRouter.
Deep-Dive Benchmarks & Teardowns
Independent benchmark analysis and real-world agent billing teardowns featuring GLM 5.3.
Compare Similar Models
Browse all models →Z.ai (GLM)
GLM 5.3 FlashX Legal
GLM-5.3-FlashX-Legal (GLM-Legal) is Z.ai's legal-domain variant of GLM-5.3-FlashX, a native multimodal model that accepts text, image, and video input and produces text. It adds legal-domain mid-training and post-training with reinforcement...
Z.ai (GLM)
GLM 5.3 Prime
GLM-5.3-Prime is the high-speed variant of Z.ai's GLM-5.3, inheriting its full capabilities while delivering 1.5–2× the output throughput through inference acceleration. It supports text input and output with a 1M-token...
Z.ai (GLM)
GLM 5.3 FlashX
GLM-5.3-FlashX is the high-speed variant of Z.ai's GLM-5.3-Flash, a native multimodal model delivering inference speeds of up to 200 tokens/s. Built on the same hybrid sparse and linear attention architecture...
Z.ai (GLM)
GLM 5.3 Flash
GLM-5.3-Flash is a native multimodal model from Z.ai. It is suited for efficient coding and long-horizon agent tasks. Its hybrid sparse and linear attention architecture maintains accurate long-context behavior while...
Anthropic
Claude Opus 5.5
Claude Opus 5.5 is Anthropic's flagship model for demanding reasoning, coding, and long-horizon agentic work, succeeding Claude Opus 5. It is particularly strong at multi-step changes in large codebases, code...
Moonshot AI
Kimi K3.1 Pro
Kimi K3.1 Pro is Moonshot AI's flagship open-weight multimodal reasoning model and a post-trained successor to Kimi K3, with a 1M-token context window. It is built for long-horizon coding, 3D...