Skip to content

GLM 5.3 (batch)

text → textHugging Face OpenRouter Cost calculator →

Best price · blended / 1M

$0.402

at Wafer

Input / output
$0.0699 / $1.40
Providers
37
Spread
3.5×
7 days
0%
30 days
−1.7%
Context
1M
Prices per 1M tokens, blended: Wafer (us) $0.402, Wafer $0.484, Featherless $0.645, DeepInfra (fp4/flex) $0.838, Decart $0.873, InferenceNet $0.963, InferenceNet (us) $0.963, Novita $0.968, SiliconFlow $1.00, DeepInfra $1.05, Cloudflare $1.075, GMICloud $1.075, Tenstorrent $1.075, Relace $1.15, AkashML $1.24, Cohere $1.25, Morph $1.29, CoreWeave $1.31, Sail Research (us) $1.31, DigitalOcean $1.40, Fireworks (spot) $1.505, Inceptron $1.64, Alibaba $1.83, Sail Research $1.84, Friendli $1.935, AtlasCloud $2.15, Baidu $2.15, BaseTen $2.15, BaseTen (fp4/us) $2.15, Crusoe $2.15, Fireworks $2.15, Mistral (zdr) $2.15, Modal $2.15, Nebius $2.15, Parasail $2.15, SambaNova $2.15, Together $2.15, Venice $2.15, Z.AI $2.15, Makora $2.225, Mistral (nvfp4/eu) $2.365, Doubleword $2.38, Alibaba (prime) $2.58, Amazon Bedrock (us-east-1) $2.58, BaseTen (fast) $3.225, BaseTen (us/fast) $3.225, Fireworks (fast) $3.225, ModelRun $3.225, Parasail (fast) $3.225.

Providers

  • WaferCheapest
    us
    $0.402
    $0.0699 in · $1.40 out
    −56% vs ref
    1M ctx100% uptimefp4
  • $0.484
    $0.179 in · $1.40 out
    −47% vs ref
    1M ctx100% uptimefp4
  • $0.645
    $0.42 in · $1.32 out
    −30% vs ref
    256K ctx96.7% uptime−70% discount
  • fp4/flex
    $0.838
    $0.45 in · $2.00 out
    −8.7% vs ref
    1M ctx100% uptimefp4·−38% discount
  • $0.873
    $0.56 in · $1.81 out
    −4.9% vs ref
    1M ctx100% uptimefp4
  • $0.963
    $0.084 in · $3.60 out
    +4.9% vs ref
    1M ctx100% uptimefp4
  • $0.963
    $0.084 in · $3.60 out
    +4.9% vs ref
    1M ctx100% uptimefp4
  • $0.968
    $0.63 in · $1.98 out
    +5.4% vs ref
    1M ctx100% uptimefp8·−55% discount
  • $1.00
    $0.70 in · $1.90 out
    +9.0% vs ref
    1M ctx99.9% uptimefp8
  • $1.05
    $0.563 in · $2.50 out
    +14% vs ref
    1M ctx99.3% uptimefp4·−38% discount
  • $1.075
    $0.70 in · $2.20 out
    +17% vs ref
    1M ctx100% uptime
  • $1.075
    $0.70 in · $2.20 out
    +17% vs ref
    1M ctx97.9% uptimefp8·−50% discount
  • $1.075
    $0.60 in · $2.50 out
    +17% vs ref
    1M ctx100% uptimefp8
  • $1.15
    $0.0703 in · $4.38 out
    +25% vs ref
    1M ctx100% uptimefp4
  • $1.24
    $0.19 in · $4.40 out
    +35% vs ref
    1M ctx100% uptimefp4
  • $1.25
    $0.50 in · $3.50 out
    +36% vs ref
    1M ctx100% uptimenvfp4
  • $1.29
    $0.671 in · $3.132 out
    +40% vs ref
    1M ctx100% uptimefp8
  • $1.31
    $0.75 in · $3.00 out
    +43% vs ref
    1M ctx100% uptimenvfp4
  • $1.31
    $0.75 in · $3.00 out
    +43% vs ref
    1M ctx100% uptimefp8
  • $1.40
    $0.91 in · $2.86 out
    +52% vs ref
    1M ctx100% uptime
  • $1.505
    $0.98 in · $3.08 out
    +64% vs ref
    1M ctx99.5% uptime
  • $1.64
    $1.05 in · $3.40 out
    +78% vs ref
    1M ctx99.9% uptimefp4
  • $1.83
    $1.19 in · $3.74 out
    +99% vs ref
    1M ctx100% uptime
  • $1.84
    $1.05 in · $4.20 out
    +100% vs ref
    1M ctx100% uptimefp8
  • $1.935
    $1.26 in · $3.96 out
    +111% vs ref
    1M ctx100% uptime−10% discount
  • $2.15
    $1.40 in · $4.40 out
    +134% vs ref
    1M ctx100% uptimefp8
  • $2.15
    $1.40 in · $4.40 out
    +134% vs ref
    1M ctxfp8
  • $2.15
    $1.40 in · $4.40 out
    +134% vs ref
    1M ctx100% uptimefp4
  • fp4/us
    $2.15
    $1.40 in · $4.40 out
    +134% vs ref
    1M ctx100% uptimefp4
  • $2.15
    $1.40 in · $4.40 out
    +134% vs ref
    1M ctx100% uptimefp4
  • $2.15
    $1.40 in · $4.40 out
    +134% vs ref
    1M ctx100% uptime
  • $2.15
    $1.40 in · $4.40 out
    +134% vs ref
    1M ctx99.8% uptimenvfp4
  • $2.15
    $1.40 in · $4.40 out
    +134% vs ref
    1M ctx99.9% uptime
  • $2.15
    $1.40 in · $4.40 out
    +134% vs ref
    1M ctx100% uptimefp4
  • $2.15
    $1.40 in · $4.40 out
    +134% vs ref
    1M ctx100% uptimefp4
  • $2.15
    $1.40 in · $4.40 out
    +134% vs ref
    1M ctx100% uptime−50% discount
  • $2.15
    $1.40 in · $4.40 out
    +134% vs ref
    1M ctx100% uptime
  • $2.15
    $1.40 in · $4.40 out
    +134% vs ref
    1M ctx100% uptime−20% discount
  • $2.15
    $1.40 in · $4.40 out
    +134% vs ref
    1M ctx100% uptimefp8
  • $2.225
    $1.50 in · $4.40 out
    +142% vs ref
    1M ctx99.1% uptimefp4
  • nvfp4/eu
    $2.365
    $1.54 in · $4.84 out
    +158% vs ref
    1M ctx99.6% uptimenvfp4
  • $2.38
    $1.36 in · $5.44 out
    +159% vs ref
    1M ctx100% uptimenvfp4
  • prime
    $2.58
    $1.68 in · $5.28 out
    +181% vs ref
    1M ctx100% uptime
  • $2.58
    $1.68 in · $5.28 out
    +181% vs ref
    1M ctx99.5% uptime
  • $3.225
    $2.10 in · $6.60 out
    +251% vs ref
    1M ctx100% uptimefp8
  • us/fast
    $3.225
    $2.10 in · $6.60 out
    +251% vs ref
    1M ctx100% uptimefp8
  • $3.225
    $2.10 in · $6.60 out
    +251% vs ref
    1M ctx99.7% uptime
  • $3.225
    $2.10 in · $6.60 out
    +251% vs ref
    1M ctx100% uptimefp4
  • $3.225
    $2.10 in · $6.60 out
    +251% vs ref
    1M ctx100% uptimefp4
Blended = (3 × input + output) / 4. Ref $0.918 = median of cheapest 20%.

Price history

USD per 1M tokens · each step is a price change

Prices in USD / 1M tokens.

Other pricing

Offered byCache readCache write
Wafer (us)$0.0689—
Wafer$0.178—
Featherless$0.078—
DeepInfra (fp4/flex), Decart +2$0.10—
InferenceNet, InferenceNet (us)$0.079—
Novita$0.117—
SiliconFlow, Cloudflare +1$0.13—
DeepInfra$0.125—
Relace$0.0703—
AkashML, CoreWeave$0.19—
Morph, Doubleword$0.136—
Sail Research (us)$0.12—
DigitalOcean$0.169—
Fireworks (spot)$0.182—
Inceptron$0.125—
Alibaba$0.238—
Sail Research$0.168—
Friendli$0.234—
AtlasCloud, Baidu +10$0.26—
BaseTen, BaseTen (fp4/us) +1$0.14—
SambaNova$0.07—
Mistral (nvfp4/eu)$0.154—
Alibaba (prime)$0.336—
Amazon Bedrock (us-east-1)$0.312$2.10
BaseTen (fast), BaseTen (us/fast)$0.21—
Fireworks (fast)$0.39—
ModelRun$0.40—

Changes

  • Cache read price,
    $0.14→$0.136−2.9%
  • Output price,
    $4.40→$5.44+23.6%
  • Input price,
    $1.40→$1.36−2.9%
  • Cache read price,
    $0.17→$0.136−20.0%
  • Output price,
    $3.915→$3.132−20.0%
  • Input price,
    $0.839→$0.671−20.0%
  • No longer served,
    Delisted
  • Now served,
    Listed
  • Cache read price,
    $0.147→$0.125−15.0%
  • Output price,
    $4.08→$3.40−16.7%
  • Cache read price,
    $0.136→$0.17+25.0%
  • Output price,
    $3.132→$3.915+25.0%
  • Input price,
    $0.671→$0.839+25.0%
  • Cache read price,
    $0.206→$0.136−34.0%
  • Output price,
    $4.747→$3.132−34.0%

About

The cheapest provider for GLM 5.3 (batch) is Wafer at $0.402 per 1M tokens (blended), charging $0.0699 per 1M input and $1.40 per 1M output tokens. 37 providers serve it; the most expensive charges 3.5× the typical price. Its best price fell 1.7% over the last 30 days. Context window: 1M tokens. Prices tracked since Oct 3, 2026.

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...
Released
Aug 18, 2026
First tracked
Oct 3, 2026
Canonical slug
z-ai/glm-5.3-20260816
Weights
zai-org/GLM-5.3